life

Review
A Chemist’s Perspective on the Role of Phosphorus at
the Origins of Life
Christian Fernández-García, Adam J. Coggins and Matthew W. Powner * ID

Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, UK;
christian.fernandez@ucl.ac.uk (C.F.-G.); a.coggins@ucl.ac.uk (A.J.C.)
* Correspondence: matthew.powner@ucl.ac.uk; Tel.: +44-020-7679-4542

Received: 28 June 2017; Accepted: 11 July 2017; Published: 13 July 2017

Abstract: The central role that phosphates play in biological systems, suggests they also played an
important role in the emergence of life on Earth. In recent years, numerous important advances have
been made towards understanding the influence that phosphates may have had on prebiotic chemistry,
and here, we highlight two important aspects of prebiotic phosphate chemistry. Firstly, we discuss
prebiotic phosphorylation reactions; we specifically contrast aqueous electrophilic phosphorylation,
and aqueous nucleophilic phosphorylation strategies, with dry-state phosphorylations that are
mediated by dissociative phosphoryl-transfer. Secondly, we discuss the non-structural roles that
phosphates can play in prebiotic chemistry. Here, we focus on the mechanisms by which phosphate
has guided prebiotic reactivity through catalysis or buffering effects, to facilitating selective
transformations in neutral water. Several prebiotic routes towards the synthesis of nucleotides,
amino acids, and core metabolites, that have been facilitated or controlled by phosphate acting as
a general acid–base catalyst, pH buffer, or a chemical buffer, are outlined. These facile and subtle
mechanisms for incorporation and exploitation of phosphates to orchestrate selective, robust prebiotic
chemistry, coupled with the central and universally conserved roles of phosphates in biochemistry,
provide an increasingly clear message that understanding phosphate chemistry will be a key element
in elucidating the origins of life on Earth.

Keywords: prebiotic chemistry; phosphate; phosphorylation; nucleotides; amino acids; general

acid-base catalyst

1. Introduction
Phosphates are essential to modern biological systems, and their wide and varied range of
biological roles is a testament to their value in controlling chemistry and building robust structures in
an aqueous environment. They provide the stable ligation required to fix information in RNA and
DNA, contribute to cellular structure in phospholipids, serve as the basic currency of biochemical
energy (e.g., ATP, phosphoenol pyruvate (PEP), creatine phosphate (CP)) (Figure 1), and feature in a
wide variety of metabolites and commonly observed post-translational protein modifications [1].
Westheimer provided a detailed analysis of the essential role of phosphate in living systems
more than 30 years ago [2], highlighting in particular, that phosphates are ionized at physiological
pH, due to a low first pKa (pKa = 2.2, 7.2, 12.3). This ionic character renders phosphates hydrophilic,
and facilitates their retention within a cell membrane. Importantly, in the case of RNA and DNA
ligation, the ionic structure of phosphates allows the ligation of two nucleosides whilst retaining a
negative charge at the phosphodiester. The charge carried by the phosphodiester ligations between
nucleotides provides an essential solubilising element, and importantly, protects the phosphodiesters
from hydrolysis. Beyond the structural role phosphates play in biology, they also serve a multitude
of integral roles in energy metabolism, where again the kinetic stability afforded by ionization is

Life 2017, 7, 31; doi:10.3390/life7030031 www.mdpi.com/journal/life

Life 2017, 7, 31 2 of 23

Life 2017, 7, 31 2 of 23
an essential element in exploiting phosphates. Kinetic stability and thermodynamic activation are
essentialtoelement
coupled excellentineffect
exploiting phosphates.
in phosphate Kinetic
moieties, stability
to provide and thermodynamic
a robust activation
chemical drive for are
biochemical
coupled to excellent effect in phosphate moieties, to provide a robust chemical drive for biochemical
transformations, whilst allowing enzymatic modulation of reactions, and regulation of metabolic
transformations,
pathways [3]. whilst allowing enzymatic modulation of reactions, and regulation of metabolic
pathways [3].

Figure 1.1.Biologically
Biologically important
important phosphates,
phosphates, exemplifying
exemplifying the central
the central role phosphates
role phosphates play in
play in biological
biological information
information transfer,and
transfer, structure, structure,
energyand energy metabolism.
metabolism.

Given the
Given the deep-seated
deep-seated role role phosphate
phosphate plays plays in in life’s
life’s most
most highly
highly conserved
conserved processes,
processes, it it is
is
essential to consider the role of phosphate at the origins of life.
essential to consider the role of phosphate at the origins of life. Numerous proposals have been Numerous proposals have been made
that the
made earliest
that stages of
the earliest evolution
stages might not
of evolution havenot
might used a phosphorus-based
have used a phosphorus-based biochemistry, including
biochemistry,
phosphate-free metabolism [4], lipids [5–8], and genetics [9–12],
including phosphate-free metabolism [4], lipids [5–8], and genetics [9–12], however, these hypotheses however, these hypotheses do not
address the fundamental question of how biology became
do not address the fundamental question of how biology became addicted to phosphate; they merely addicted to phosphate; they merely
postpone this
postpone this question.
question. Therefore,
Therefore, we we will
will focus
focus here
here on on thethe chemistry
chemistry of of phosphate,
phosphate, and and specifically
specifically
P(V) phosphorus chemistry, which is the cornerstone
P(V) phosphorus chemistry, which is the cornerstone of extant metabolism. We do of extant metabolism. We do not aim to review
not aim to
all phosphate chemistry relevant to the origins of life; rather, we
review all phosphate chemistry relevant to the origins of life; rather, we seek to highlight a few seek to highlight a few key elements
of phosphate
key elements of chemistry
phosphate that we havethat
chemistry foundwe haveespecially
foundinstructive in our investigations
especially instructive of prebiotic
in our investigations of
chemistry.
prebiotic chemistry.
At this
At this juncture,
juncture, it it must
must be be noted
noted that
that the the availability
availability of of inorganic
inorganic phosphate
phosphate on on the
the early
early Earth
Earth
has been widely debated. Orthophosphate (PO 43−
3 −), the
has been widely debated. Orthophosphate (PO4 ), the most common form of phosphorus on themost common form of phosphorus on the
Earth [13], is largely found as apatite minerals, which are relatively
Earth [13], is largely found as apatite minerals, which are relatively insoluble in most geological insoluble in most geological
settings. The
settings. The insolubility
insolubility of of apatite
apatite minerals
minerals has has been
been referred
referred to to as
as the
the ‘phosphate
‘phosphate problem’
problem’ [14–16],
[14–16],
and of course, useful production of soluble phosphate requires
and of course, useful production of soluble phosphate requires that this phosphate remains that this phosphate remains soluble
soluble
long enough,
long enough, and and inin sufficient
sufficient quantity,
quantity, to to bebe utilised
utilised in in prebiotic chemistry. This
prebiotic chemistry. This may
may imply
imply aa specific
specific
environment is required for exploitation of phosphate. For
environment is required for exploitation of phosphate. For example, one might consider an example, one might consider an aqueous
aqueous
environment low
environment low in in soluble
soluble calcium,
calcium, whichwhich wouldwould be be expected
expected to to rapidly
rapidly precipitate
precipitate orthophosphate,
orthophosphate,
might be essential to exploit phosphate at the origins of life, and
might be essential to exploit phosphate at the origins of life, and that the key role of phosphates that the key role of phosphates in in
life (and prebiotic chemistry) suggests that the liberation of phosphate
life (and prebiotic chemistry) suggests that the liberation of phosphate by other bulk anions may by other bulk anions may have
played
have a keya key
played role role
in the
in theorigins
origins of oflife.
life.TheThechallenges
challengesimposed imposed by by geochemically accessing
geochemically accessing
phosphorus in
phosphorus in the
the environment
environment are are certainly
certainly not not trivial,
trivial, but but these
these constraints
constraints suggest
suggest it it is
is likely
likely that
that
phosphate (and its availability) could play a pivotal role in bringing
phosphate (and its availability) could play a pivotal role in bringing together prebiotic chemistry and together prebiotic chemistry and
early Earth
early Earth geochemical
geochemical models, models, as as well
well as as aa pivotal
pivotal role role in in the
the origins
origins of of life
life itself. Severalreviews
itself. Several reviews
have discussed the availability of phosphorus on the early Earth,
have discussed the availability of phosphorus on the early Earth, and numerous attempts to quantify and numerous attempts to quantify
phosphorus availability
phosphorus availabilityon onthetheprimitive
primitiveEarth Earthhave havebeen beenmade made(Scheme
(Scheme 1)1)[17–23].
[17–23]. WeWedodo notnot
intend
intendto
recover
to recover this ground
this ground here;
here;whilst
whilst thetheissue
issueofofphosphate
phosphateavailability
availabilitywill willcontinue
continuetotostimulate
stimulate debate
debate
within the field of prebiotic chemistry, it is clear that multiple mechanisms
within the field of prebiotic chemistry, it is clear that multiple mechanisms for the accumulation of for the accumulation of
useful phosphates under specific conditions on the early Earth are
useful phosphates under specific conditions on the early Earth are at least plausible. Indeed, the total at least plausible. Indeed, the
total amount of soluble phosphate in the entire hydrosphere need not be high if prebiotic phosphorus
chemistry was confined to specific (phosphorus-rich) niches [16,17,24]. Accordingly, effective and
selective mechanisms for phosphate incorporation into organic molecules may pose a greater
Life 2017, 7, 31 3 of 23

amount of soluble phosphate in the entire hydrosphere need not be high if prebiotic phosphorus
chemistry was confined to specific (phosphorus-rich) niches [16,17,24]. Accordingly, effective and
selective mechanisms
Life 2017, 7, 31 for phosphate incorporation into organic molecules may pose a greater challenge
3 of 23
to prebiotic chemistry than the global phosphorus inventory, and here, we will focus upon illustrative
challenge
examples to prebioticchemistry
of phosphate chemistry than the global
in prebiotic phosphorus inventory, and here, we will focus upon
reactions.
illustrative examples of phosphate chemistry in prebiotic reactions.

Scheme
Scheme 1. An
1. An overview
overview ofof suggestedprebiotic
suggested prebiotic routes
routes to
to orthophosphate
orthophosphateandandcondensed
condensed phosphates
phosphates
production (black arrows), phosphate interconversion (green arrows), phosphate solubilisation
production (black arrows), phosphate interconversion (green arrows), phosphate solubilisation (red
arrows), and urea-mediated phosphorylation (blue arrows). N = nucleoside, pN = nucleotide,
(red arrows), and urea-mediated phosphorylation (blue arrows). N = nucleoside, pN = nucleotide, cTMP
= cyclotriphosphate.
cTMP = cyclotriphosphate.

2. Phosphorylation in Water
2. Phosphorylation in Water
Electrophilic phosphorylation reactions in water require a high degree of selectivity, for
Electrophilic
example, phosphorylation
the phosphorylation of areactions
(nucleotide) in water require
hydroxyl a high
moiety degree
in water of selectivity,
requires for example,
direct competition
the with
phosphorylation of a (nucleotide) hydroxyl moiety in water requires direct
solvent (55 M water), which is (weakly) nucleophilic, and can be phosphorylated to yield competition with
solvent (55 Mphosphate.
inorganic water), which is (weakly)
The most nucleophilic,
widely investigated and cantobe
approach phosphorylated
prebiotic to yield
phosphorylation inorganic
has been
phosphate. The most
by electrophilic widely
activation investigated approach
of orthophosphate toactivated
to generate prebiotic(anhydride-type)
phosphorylation has been by
intermediates,
which can activation
electrophilic in turn reactofwith nucleophiles (e.g.,
orthophosphate hydroxyl groups)
to generate activated to afford phosphorylated
(anhydride-type) products
intermediates,
(e.g., phosphate esters). numerous activating agents have been investigated,
which can in turn react with nucleophiles (e.g., hydroxyl groups) to afford phosphorylated productsand cyanoacetylene (1)
[25], cyanogen (2) [26], cyanamide (3) [27], and cyanate (4) [28], are all noted
(e.g., phosphate esters). numerous activating agents have been investigated, and cyanoacetylene worthy examples of
prebiotically
(1) [25], cyanogen plausible
(2) [26],electrophiles
cyanamide (3) that[27],
haveandbeen exploited
cyanate in this
(4) [28], role.
are all However,
noted worthyaqueous
examples
electrophilic phosphorylation reactions can suffer dramatically from adverse competition with water,
of prebiotically plausible electrophiles that have been exploited in this role. However, aqueous
and typically aqueous phosphorylations are low yielding. For example, phosphorylation of 0.16M
electrophilic phosphorylation reactions can suffer dramatically from adverse competition with water,
uridine (5) with cyanoformamide (6) in pH 8 1M phosphate solution furnishes only 1% – 4% uridine-
and typically aqueous phosphorylations are low yielding. For example, phosphorylation of 0.16
5’-phosphate (UMP) (Scheme 2) [25,28]. Incubation of sugars (for example, ribose (7)) with cyanogen
M uridine (5) with cyanoformamide (6) in pH 8 1 M phosphate solution furnishes only 1–4%
(2) or cyanamide (3) in phosphate solution, also affords phosphorylated sugars (for example, ribose-
uridine-5’-phosphate
1-phosphate (R1P)),(UMP) (Scheme
but again with low 2) yield
[25,28]. Incubation
(10%–20%) of sugars (for example, ribose (7)) with
[26,27].
cyanogen (2) or cyanamide (3) in phosphate solution, also affords phosphorylated sugars (for example,
ribose-1-phosphate (R1P)), but again with low yield (10–20%) [26,27].
Life 2017, 7, 31 4 of 23
Life 2017, 7, 31 4 of 23

Scheme
Scheme 2. Phosphorylation
2. Phosphorylation of of
uridine
uridine(5)(5)
andandribose
ribose(7)
(7)using
usingorthophosphate
orthophosphateand
and cyanoformamide
cyanoformamide (6),
cyanogen (2), or (2),
(6), cyanogen cyanamide (3), in(3),
or cyanamide aqueous solution.
in aqueous solution.

The soluble phosphates obtained by the oxidation of schreibersite, have been considered for
The soluble phosphates obtained by the oxidation of schreibersite, have been considered for
localised delivery of phosphate [29], and schreibersite has been used as a direct phosphorus source
localised delivery of phosphate [29], and schreibersite has been used as a direct phosphorus source for
for electrophilic phosphorylation. For example, aqueous solutions of glycerol and schreibersite
electrophilic phosphorylation. For example, aqueous solutions of glycerol and schreibersite incubated
incubated at 65 °C under anaerobic conditions afforded glycerol phosphate (2.5%) [30], and minerals
◦ C under anaerobic conditions afforded glycerol phosphate (2.5%) [30], and minerals related
at 65
related to schreibersite, such as Fe3P and Fe2NiP, when combined with nucleosides in aqueous solution, also
to schreibersite,
return mixturessuch as Fe3 Pinand
of nucleotides low Fe 2 NiP,
yield when
(up to 1% – combined with
6% of the total nucleosides
dissolved in aqueous
phosphorus) solution,
[31]. However,
alsomodel
returnstudies
mixtures
of the aqueous oxidation of Fe3P (used as a model for schreibersite), affords significant[31].
of nucleotides in low yield (up to 1–6% of the total dissolved phosphorus)
However, model
quantities studies of the
of orthophosphate andaqueous oxidation
pyrophosphate, of Fe3minor
alongside P (used as a model
reduced for schreibersite),
phosphorus affords
species [20–22].
significant
The amount quantities of orthophosphate
of condensed and pyrophosphate,
phosphates returned alongside
have been increased minor peroxide
by hydrogen reduced oxidation
phosphorus
of the
species reduced-phosphorus
[20–22]. The amount of species corroded
condensed from schreibersite
phosphates returned[23],
haveto been
affordincreased
a good yield (up to
by hydrogen
34%) of condensed phosphates, including pyrophosphate, triphosphate, and
peroxide oxidation of the reduced-phosphorus species corroded from schreibersite [23], to afford cyclotriphosphate
(cTMP);
a good yieldthe(up
resttoof34%)
the phosphorus
of condensed is returned as orthophosphate.
phosphates, including pyrophosphate, triphosphate, and
cyclotriphosphate (cTMP); the rest of the phosphorus is returned as orthophosphate.
3. Phosphorylation in Water with Polyphosphates
3. Phosphorylation in Water
Polyphosphates play a with Polyphosphates
dominant role in biological activation and phosphorylation; they are also
intrinsically
Polyphosphates activated
play aand generally
dominant role inmore soluble
biological than and
activation orthophosphate.
phosphorylation; Consequently,
they are also
phosphorylation reactions exploiting condensed phosphates have drawn significant attention.
intrinsically activated and generally more soluble than orthophosphate. Consequently, phosphorylation
Numerous attempts have been made to convert orthophosphate to polyphosphates, and pyrophosphate
reactions exploiting condensed phosphates have drawn significant attention. Numerous attempts have
and triphosphate synthesis has been achieved by various condensing agents in water [32–35], but perhaps
been made to convert orthophosphate to polyphosphates, and pyrophosphate and triphosphate
the simplest conditions for polyphosphate synthesis are heating phosphates (H3PO4, NaH2PO·H2O,
synthesis
NH4H2PO has4, etc.)
beenin achieved byatvarious
the dry-state condensing
high temperature (80 –agents
160 °C).inDry-heating
water [32–35], but perhaps
orthophosphate salts the
simplest conditions for polyphosphate synthesis are heating phosphates
furnishes pyrophosphate (5% – 50%), and triphosphate (1%–30%), in 2 hours at 160 °C [36], and even (H 3 PO 4 , NaH 2 PO ·H2 O,
NHat H PO , etc.) in the dry-state at high temperature (80–160 ◦ C). Dry-heating orthophosphate
4 low
2 temperature
4 (37 °C), insoluble calcium and magnesium orthophosphates have been condensed
saltsto furnishes pyrophosphate ◦ C [36], and
afford pyrophosphate (albeit (5–50%), and [37].
in 0.06% yield) triphosphate
Condensed(1–30%),
phosphates in have
2 h also
at 160
been produced
even ◦
byat low temperature
magmatic processing (37 C), insoluble
of mixtures calcium
of apatite and[38],
and basalt magnesium orthophosphates
and the simplicity and robustnesshave of been
condensed to afford pyrophosphate (albeit in 0.06% yield) [37]. Condensed phosphates have also
polyphosphate synthesis strongly suggests polyphosphates may have played an important role in
beentheproduced
origin of life,by as well as extant
magmatic biology.of
processing Moreover,
mixturesurea (8), a simple,
of apatite and highly
basalt prebiotically
[38], and the plausible
simplicity
andcompound,
robustness has been demonstrated
of polyphosphate synthesis to strongly
significantly improve
suggests phosphate polymerisation.
polyphosphates may have played
Orthophosphate
an important role has beenorigin
in the converted to polyphosphates
of life, as well as extantin highbiology.
yield (60%) at moderate
Moreover, urea temperatures
(8), a simple,
(72 °C in 26 days) in the presence of urea (8), and under similar
highly prebiotically plausible compound, has been demonstrated to significantly improve conditions, in the presence of
phosphate
nucleosides, cyclotrimetaphosphate (cTMP) has been obtained in 23% yield [39]. The most facile
polymerisation. Orthophosphate has been converted to polyphosphates in high yield (60%) at
transformations have been observed with NH4H2PO4, and the related mineral struvite (MgNH4PO4
moderate temperatures (72 ◦ C in 26 days) in the presence of urea (8), and under similar conditions,
6H2O), which can be formed from soluble phosphates when ammonium concentrations exceed 0.01
in the presence of nucleosides, cyclotrimetaphosphate (cTMP) has been obtained in 23% yield [39].
M [40], and appears to be kinetically favoured over the more thermodynamically stable
Thehydroxyapatite
most facile transformations
[41]. Struvite have beenexploited
has been observedinwith the NH 4 H2 PO4 , and the
phosphorylation related mineral
of nucleosides [40], struvite
and
(MgNH 4 PO 4 · 6H 2 O), which can be formed from soluble phosphates
pyrophosphate can be obtained from struvite in up to 88% after 48 hr at 85 °C [42]. when ammonium concentrations
exceed 0.01Linear M [40], and appearssuch
polyphosphates, to be
askinetically favoured over
sodium triphosphate, and the more thermodynamically
adenosine (9), afford mixtures stable of
hydroxyapatite
adenosine 2’, [41]. 3’-, andStruvite has beeninexploited
5’-phosphates low yield in thewhen
(1%) phosphorylation of nucleosides
they are refluxed [40], and
in basic aqueous
pyrophosphate can be obtained from struvite in up to 88% after 48 h at 85 ◦ C [42].
Life 2017, 7, 31 5 of 23

Linear polyphosphates, such as sodium triphosphate, and adenosine (9), afford mixtures of
adenosine 2’, 3’-, and 5’-phosphates in low yield (1%) when they are refluxed in basic aqueous
solutions for short periods (4–6 h). Though pyrophosphate is not observed to afford nucleotides under
similar conditions, longer polyphosphates, such as cyclic hexametaphosphate, afford the same set
of products as triphosphate, and in comparable yield [43]. However, of the various polyphosphates
examined, cyclotrimetaphosphate (cTMP) is of particular note (Scheme 3). Incubation of sodium
cTMP with 9 at high pH affords a mixture of 2’- and 3’-phosphates (9-2p and 9-3p; ~1:1) in 31%
yield. It is also of note that phosphorylation of 2’-deoxyadenosine (10), is an order of magnitude
less efficient under similar conditions, and only affords a mixture of 50 - and 3’-phosphates in 2%
yield [44]. Moreover, phosphorylation of adenosine-5’-phosphate (AMP) with cTMP at 100 ◦ C
yields only small quantities of nucleotide polyphosphates (ADP and ATP in 0.03% and 0.09% yield,
respectively), accompanied by significant decomposition to adenosine (9) and free base (adenine) [45].
These results make clear the importance of the vicinal diol in efficient cTMP-phosphorylation of
ribonucleotides. The cTMP-mediated phosphorylation of 9 can also be performed at neutral pH,
if Mg2+ is present, however the reaction is sluggish and the absolute selectivity for diol phosphorylation
is lost, affording a mixture of 20 ,30 -cyclicphosphate 11 (3.8%) and AMP (<1%) [46]. Interestingly,
however, exploiting a mixture of tetramethylammonium and sodium counter ions with cTMP results
in drastically improved reactivity, and furnishes a mixture of 2’- and 3’-phosphates in 70–90% yield
(at high pH) [47]. Nucleotide-5’-phoshates were not detected in these experiments, demonstrating
the remarkable selectivity for diol-phosphorylation in water. The suspected mechanism for these
high yielding reactions involves specific base-catalysed phosphorylation of vicinal diol by cTMP
to yield the nucleotide 2’- or 3’-triphosphate. The significantly lowered pKa of the diol results in a
highly regioselective triphosphorylation, and the close proximity of the second alcohol moiety of
the diol then results in attack at the α-phosphorus to form a 2’,3’-cyclic phosphate, whilst liberating
pyrophosphate. The strained cyclic phosphate then undergoes alkaline hydrolysis to afford a mixture
of 9-2p and 9-3p in excellent yield (Scheme 3). Though nucleotide 5’-phosphates are universally
exploited in extant biology, it is of note that RNA hydrolysis proceeds via the transient formation
of 2’,3’-cyclic phosphates. Therefore, it also appears likely that prebiotic RNA recycling would
exploit 2’,3’-cyclic phosphates, and 2’,3’-cyclic phosphates would provide the most direct entry point
for the advent of continuous (prebiotic) RNA evolution strategies. Though these cTMP-mediated
phosphorylations require preformed ribonucleoside, they are highly instructive for understanding the
value of 2’,3’-phosphates during the stepwise synthesis of ribonucleotides, and we will return to the
chemistry of nucleotide-2’,3’-phosphates in due course.
Phosphorylations with cTMP have found many applications beyond prebiotic nucleotide
synthesis. For example, small (simple) oligopeptides can be readily obtained by cTMP-mediated
coupling of amino acids in aqueous solution [48–50]. Incubation of glycine (Gly) with cTMP affords
diglycine in 35% yield at 70 ◦ C after 70 h (Scheme 4); even at room temperature, the dipeptide
is obtained in moderate yield (20–22%) after 190 h. Alanine (Ala) and aspartic (Asp) acid have
also been observed to afford their corresponding dipeptides, however, the yields (12% and 2–5%,
respectively) [48,49] are significantly suppressed relative to diglycine [50]. Serine (Ser), on the
other hand, has only been observed to undergo O-phosphorylation, to afford small amounts of
O-phosphoserine (12; 4%) as the only product [49,51].
The proposed mechanism for cTMP activation of amino acids occurs by a two-step mechanism
(Scheme 5). First, nucleophilic attack of the amine moiety at phosphorus results in the formation of an
N-triphosphoramidate 13, then formation of a five-membered cyclic mixed anhydride intermediate
(phosphoramidate 14) occurs by attack of the carboxylate on the same phosphorus, coupled to
displacement of pyrophosphate. Nucleophilic attack at carbon of the (activated) mixed anhydride 14,
by the amine moiety of another amino acid, then affords a N-phosphoro-dipeptide and dipeptide upon
phosphoramide hydrolysis [52,53]. It is of note that this mechanism proceeds via a cyclic intermediate
akin to the phosphorylation of the vicinal diol of a nucleotide.
Life 2017, 7, 31 6 of 23
Life 2017,
Life 2017, 7, 31 6 of
of 23
23
Life 2017,7,7,31
31 66 of 23

N O O
O O N
N
N O O N
N
O
O N NH 2 P
P O N NH 2
O N
N NH
NH22 O P O
O O O N
N NH
NH 22
RO O O R==H
H HO
RO
RO O PP P OO R HO
N N
N +
+ O P O P
O P O N N
N
N N + N N
HO
HO OH N
OH
O O
OO
O O O
O O O O
O
OH
OH
OH
N
HO OH O O
99R=H
9 R=H
R=H 2- cTMP P P P
cTMP
cTMP O P O O P O
O P O
AMP R=PO
R=PO3332-2- O
AMP
AMPR=PO O
O O O O
O
O
(+ 2'-triphosphate)
(+ 2'-triphosphate)
RR=
==PO
PO 2-
R PO3332-2-

NN
N
O
O
O NN NH
N NH222
NH
RO
RO
RO
NN N
N N
N
N
N N
HO OH O
O N NH 2
NH
NH
HO
HO OH
OH N 22
3-
HO
HO
ADP
ADPRRR===PPP2OO63- 3- N N
N + PP
PP i
ADP 22O66 N
N N ++ PP ii
ATP RR===PPP3O O 4-
4- O O O
ATPR
ATP O94- O
33 99 P
P
++ O
O O O
+
NN
N 11
11
O
O
O NN NH
NH222
N NH
HO
HO
HO
NN N
N
N Hydrolysis
N Hydrolysis
Hydrolysis
HO
HO
HO OH
OH
OH
99
9
N
N
N
++
+ NH22
O
O N
N NH
NH
N 2
NH
NH222
NH HO
HO N
NN N
N N
N
N NN N
N R
R11O
O OR22
OR
OR 2
NN
N NN
HH N 2-
H 9-2p R
9-2p R11 =
= H,
H, R = PO
R222 =
= PO332-
PO 2-
3
Adenine
Adenine 9-3p R = PO 2-, R = H
2-
Adenine 9-3p R1 = PO332-, R
1 3 R22 == H
H2

Scheme
Scheme
Scheme 3.3.Selective
Selective
Selective
3. Selective
Scheme 3. and
andand high
high
high
and high yielding
yielding
yielding nucleoside
yieldingnucleoside
nucleoside phosphorylation
nucleoside phosphorylation
phosphorylation by cyclotrimetaphosphate
cyclotrimetaphosphate
byby
cyclotrimetaphosphate (cTMP).
(cTMP).
cyclotrimetaphosphate (cTMP).
(cTMP).

Scheme
Scheme
Scheme
Scheme 4.4. Ligation
4. Ligation
4. Ligation
Ligation of glycine
of glycine
of
of glycine
glycine (Gly), alanine
(Gly), alanine
(Gly),
(Gly), (Ala), and
alanine aspartic
(Ala),
(Ala), aspartic
acid (Asp)
and aspartic
and acid
acid (Asp)
to afford
(Asp) to
to afford
(Asp) their
to afford
afford their
corresponding
their
their
corresponding
corresponding
dipeptides dipeptides
dipeptides
under aqueous
corresponding dipeptides under
under
conditions aqueous
aqueous
promoted
under aqueous conditions promoted by
by cyclotrimetaphosphate
conditions cyclotrimetaphosphate
cyclotrimetaphosphate (cTMP).
(cTMP).
(cTMP). Serine (cTMP).
promoted by cyclotrimetaphosphate (Ser) affords
Serine(Ser)
Serine (Ser)affords
the O-phosphorylatedaffords the
the O-phosphorylated
O-phosphorylated
product 12. product 12.
12.
Serine (Ser) affords the O-phosphorylated product

Scheme 5. Two-step mechanism for amino acid activation and peptide ligation mediated by
Scheme 5. 5. Two-step
Two-step mechanism
mechanism for
for amino
amino acid
acid activation
activation and
and peptide
peptide ligation
ligation mediated
mediated by
by
Scheme5. Two-step mechanism
Schemecyclotrimetaphosphate
cyclotrimetaphosphate (cTMP). for amino acid activation and peptide ligation mediated by
(cTMP).
cyclotrimetaphosphate
cyclotrimetaphosphate (cTMP).
(cTMP).

The mechanism for cTMP-mediated peptide activation specifically favours the activation of
(simple proteinogenic) amino acids, such as glycine, whereas β-amino acids, such as β-alanine 15,
Life 2017, 7, 31 7 of 23

Life 2017, 7, 31 7 of 23
that are widely viewed to be prebiotic, but not found in the proteome, are not observed to
be ligatedThe by mechanism
cTMP. Even for incubation
cTMP-mediated of β-alanine N-triphosphoramidate
peptide activation specifically favours (16)thedoes not result
activation of in
(simple proteinogenic) amino acids, such as glycine, whereas β-amino
activation [54,55], but upon acidification, cyclisation of the amidotriphosphate occurs, to yield acids, such as β-alanine 15, that
are widely viewed to(cTMP)
cyclotrimetaphosphate be prebiotic, butItnot
[56,57]. is found
thought in the
thatproteome, are not observed
the homologous β-alanine to be
(15)ligated
has aby higher
cTMP. Even incubation of β-alanine N-triphosphoramidate (16) does not
activation energy associated with cyclisation, and the six-membered mixed anhydride is therefore result in activation [54,55], but
upon acidification, cyclisation of the amidotriphosphate occurs, to yield cyclotrimetaphosphate
kinetically prohibited [58]. This mode of α-amino acid selective ligation could be a key mechanism
(cTMP) [56,57]. It is thought that the homologous β-alanine (15) has a higher activation energy
to select for the proteinogenic amino acids found in biology, however, it may also provide clues to
associated with cyclisation, and the six-membered mixed anhydride is therefore kinetically
why serine is ineffectual,
prohibited givenofthat
[58]. This mode the hydroxyl
α-amino moiety
acid selective of serine
ligation couldisbeideally positionedto
a key mechanism toselect
intercept
for the
N-triphosphoramidate,
the proteinogenic amino acids found in biology, however, it may also provide clues to why serinethough
to prevent activation by cyclic phosphoramidate formation. Moreover, is
appreciable yields
ineffectual, of diglycine
given (35%) can
that the hydroxyl be formed
moiety of serineby cTMP activation,
is ideally positionedfurther oligomerization
to intercept the N- to
yieldtriphosphoramidate,
triglycine is only observed
to preventto occur inby
activation very low
cyclic yield (<1%). Aformation.
phosphoramidate marginally improved
Moreover, thoughyield of
appreciable
oligoglycines yields
can of diglycine
be obtained by(35%) can be
stepwise formed when
ligation; by cTMP activation,
diglycine further oligomerization
is incubated with cTMPto under
yield
neutral or triglycine is only conditions,
slightly acidic observed to occur
moderatein very low yield
heating ◦ C) yields
(38 (<1%). A marginally
mixtures improved yield of and
of tetraglycine
oligoglycines
hexaglycine (15%can andbe4%,obtained by stepwise
respectively). ligation; when
In contrast to thediglycine
reactioniswith incubated
aminowith acidcTMP under the
monomers,
neutral or slightly acidic conditions, moderate heating (38 °C) yields mixtures of tetraglycine and
formation a cyclic phosphoramidate is not possible; a plausible mechanism for dimer ligation involves
hexaglycine (15% and 4%, respectively). In contrast to the reaction with amino acid monomers, the
formation of linear acyl-O-triphosphate 17 from the zwitterionic dipeptide, followed by dipeptide
formation a cyclic phosphoramidate is not possible; a plausible mechanism for dimer ligation
ligation. It is formation
involves of note that different
of linear pH conditions17
acyl-O-triphosphate arefromrequired for cTMP-mediated
the zwitterionic monomer
dipeptide, followed by and
dimerdipeptide
ligation,ligation.
and even It is of note that different pH conditions are required for cTMP-mediated has
then, dimer ligation is not efficient. The amine moiety of the dipeptide
a lower pKa than
monomer and the
dimeramine moiety
ligation, of thethen,
and even monomer (8.1 vs.is 9.6),
dimer ligation therefore
not efficient. Theitamine
is likely the formation
moiety of the
N-triphosphate
of thedipeptide has a lower is more
pKa thanfacile
the for
amine themoiety
dipeptide
of thethanmonomerthe monomer
(8.1 vs. 9.6),under the itconditions
therefore is likely of
thesethe formation ofPhosphorylation
experiments. the N-triphosphate ofisthe
more
dimer facile for the
does not dipeptide than the monomer
result in cyclisation under activate
(and cannot the
conditions of these experiments. Phosphorylation of the dimer does
the carboxylate moiety), therefore, this would account for the especially low yields of tripeptide not result in cyclisation (and
in thecannot activate
reaction the carboxylate
of glycine, moiety),
diglycine, and therefore,
cTMP this wouldpH.
at basic account for the especially
Significant low yields of
cTMP-phosphorylation
tripeptide in the reaction of glycine, diglycine, and cTMP at basic pH. Significant cTMP-
of the dipeptide (to afford the N-triphosphate-glygly (18)) blocks further reaction with the cyclic
phosphorylation of the dipeptide (to afford the N-triphosphate-glygly (18)) blocks further reaction with
acylphosphoramidate 14 that is required to afford trimer (Scheme 6) [59]. Although the efficient ligation
the cyclic acylphosphoramidate 14 that is required to afford trimer (Scheme 6) [59]. Although the efficient
of peptides
ligation with cTMP seems
of peptides limitedseems
with cTMP to glycine
limited dimer (GlyGly),
to glycine the (GlyGly),
dimer intramolecular phosphorylation
the intramolecular
of thephosphorylation
carboxylate group of theof glycine (Gly)
carboxylate groupforeshadowed
of glycine (Gly)the remarkable
foreshadowed theuse of amineuse
remarkable catalysis
of aminein the
phosphorylation
catalysis in theofphosphorylation
sugars by cTMP. of sugars by cTMP.

Scheme
Scheme 6. Reaction
6. Reaction of diglycine
of diglycine (GlyGly)with
(GlyGly) with cTMP
cTMP toto yield
yieldtetraglycine
tetraglycine(GlyGlyGlyGly).
(GlyGlyGlyGly). Under
Under
strongly alkaline reaction conditions, the sole product is N-triphosphate 18 and no ligation reaction
strongly alkaline reaction conditions, the sole product is N-triphosphate 18 and no ligation reaction
occurs, whereas at neutral or slightly acidic reaction conditions O-triphosphate 17 is obtained, which
occurs, whereas at neutral or slightly acidic reaction conditions O-triphosphate 17 is obtained, which
can be used to ligate the tetrapeptide GlyGlyGlyGly.
can be used to ligate the tetrapeptide GlyGlyGlyGly.
4. Phosphorylation in Water with Amine Catalysis
4. Phosphorylation in Water with Amine Catalysis
Glyceric acid 2- and 3-monophosphate 19-2p and 19-3p, which are key intermediates of glycolysis,
can be readily
Glyceric obtained
acid 2- in up to 40% yield 19-2p
and 3-monophosphate in alkaline cTMP-mediated
and 19-3p, phosphorylation
which are [60], or
key intermediates of under
glycolysis,
can be readily obtained in up to 40% yield in alkaline cTMP-mediated phosphorylation [60], or under
slightly acidic conditions (pH ≥ 6.0), when facilitated by charged-mediated absorption into a
Life 2017, 7, 31 8 of 23

Life 2017, 7, 31 8 of 23
mineral bilayer (Scheme 7) [61]. The reaction of glucose and sucrose with cTMP under alkaline
(pHslightly acidic conditions
> 13) conditions has been(pHobserved
> 6.0), when facilitated
to afford by charged-mediated
saccharide absorption intoderivatives
mono- and tri-phosphate a mineral at
room temperature [62]. However, glycolaldehyde phosphate (GCP) [63–65], undergoes(pH
bilayer (Scheme 7) [61]. The reaction of glucose and sucrose with cTMP under alkaline >13)
homoaldol
conditions has been observed to afford saccharide mono- and tri-phosphate
condensation under the alkaline conditions required for phosphorylation, with cTMP hampering derivatives at room
temperature
these reactions [64].[62].Similar
However, glycolaldehyde
problems have also beenphosphate (GCP) for
encountered [63–65], undergoes
the synthesis homoaldol
of glyceraldehyde
condensation under the alkaline conditions required for phosphorylation,
phosphate derivatives from glyceraldehyde (GA) by cTMP-mediated phosphorylation, and the with cTMP hampering these
former
reactions [64]. Similar problems have also been encountered for the synthesis of glyceraldehyde
undergoes facile (E1cB) elimination under basic conditions. However, these problems can be readily
phosphate derivatives from glyceraldehyde (GA) by cTMP-mediated phosphorylation, and the former
ameliorated by amine catalysis. The reaction of ammonia with cTMP yields amidotriphosphate
undergoes facile (E1cB) elimination under basic conditions. However, these problems can be readily
(AmTP) in excellent yield, and the amido group of AmTP is capable of reversible imine formation
ameliorated by amine catalysis. The reaction of ammonia with cTMP yields amidotriphosphate
with carbonyl
(AmTP) groups.yield,
in excellent The reversible
and the amido capture
groupofofAmTP
AmTP tethers
is capabletheofactivated
reversible phosphate to the
imine formation sugar
with
substrates,
carbonyland onceThe
groups. AmTP is tethered
reversible to the
capture of anomeric
AmTP tethers carbon,the the α-phosphate
activated is intramolecularly
phosphate to the sugar
delivered to the α-hydroxyl with exceptional control. This reactivity was exploited
substrates, and once AmTP is tethered to the anomeric carbon, the α-phosphate is intramolecularly by Krishnamurthy
et al. in the synthesis
delivered of glycolaldehyde
to the α-hydroxyl phosphate
with exceptional control.(GCP) [64]. Further
This reactivity development
was exploited of the strategy
by Krishnamurthy
wasetundertaken by Mullen
al. in the synthesis and Sutherland
of glycolaldehyde [66], (GCP)
phosphate who demonstrated that β-hydroxy-n-alkylamines
[64]. Further development of the strategy was
undertaken
react readily with by Mullen
cTMP to and Sutherland
form [66], who
amphiphiles. demonstrated
Interestingly, theythat β-hydroxy-n-alkylamines
observed that a hydrophobic react
effect
readily with cTMP to form amphiphiles. Interestingly, they observed that a hydrophobic
dictated the product distribution of this reaction, such that short-chain β-hydroxy amines afford only effect dictated
the product distribution
phosphoramide products, but of intramolecular
this reaction, such that and
transfer short-chain β-hydroxy amines
stable amphiphilic affordesters
phosphate only are
phosphoramide
obtained products,
from long-chain but intramolecular
β-hydroxy amines. transfer and stable amphiphilic phosphate esters are
obtained from long-chain β-hydroxy amines.
O
a) OH OH
P
O OH cTMP O O O O O
O + O
O P
O O O
O OH
19 19-2p 19-3p
b)
O O
P O O O O
O O
NH3 P P P NH3 P
O P P O O O O NH2 O NH
O O O O NH2 2
O O
cTMP AmTP DAP

c)
O O O OH
O AmTP P P P OH
O O O N
OH O O O H
R
R Mg2+
GC R = H
-PPi
GA R = CH2OH R OH

O O NH
OPO32- P
HO O
R
GCP R = H
GA2P R = CH2OH
d)
O OH O OH O
AmTP O O
HO HO + HO
R2 P
HO R1 O O
O O HO
P
7 R1 = OH, R2 = H O O 20
22 R1 = H, R2 = OH 21

Scheme
Scheme 7. 7.(a)(a)Cyclotrimetaphosphate
Cyclotrimetaphosphate (cTMP)
(cTMP)mediated
mediated phosphorylation of glyceric
phosphorylation acid (19).
of glyceric acid (b)(19).
(b) Ammonolysis
Ammonolysisofof cyclotrimetaphosphate
cyclotrimetaphosphate (cTMP)
(cTMP) to toobtain
obtainamidotriphosphate
amidotriphosphate (AmTP)
(AmTP) andand
diamidophosphate
diamidophosphate (DAP).
(DAP). (c) (c) Proposed
Proposed pathway
pathway for for
the the synthesis
synthesis of glycolaldehyde
of glycolaldehyde phosphate
phosphate (GCP)
(GCP) from glycolaldehyde (GC) and AmTP. (d) Synthesis of ribose phosphates
from glycolaldehyde (GC) and AmTP. (d) Synthesis of ribose phosphates (20 and 21) by reaction (20 and 21) by of
reaction of ribose (7) and AmTP.
ribose (7) and AmTP.

Eschenmoser’s tethered α-phosphorylation strategy has been extended to the syntheses of

Eschenmoser’s
several tethered Glyceraldehyde
sugar phosphates. α-phosphorylation strategy
(GA) can be has been extended
converted to the
specifically syntheses of several
to glyceraldehyde-2-
sugar phosphates. Glyceraldehyde (GA) can be converted specifically to glyceraldehyde-2-phosphate
Life 2017, 7, 31 9 of 23

Life 2017, 7, 31 9 of 23
(GA2P), but the tetrose and pentose sugars (such as ribose (7)) react to afford 1,2-cyclic, and if the
phosphatemoieties
2,3-hydroxyl (GA2P), arebutcis-disposed,
the tetrose and thenpentose sugars (such
also 2,3-cyclic as ribose Therefore,
phosphates. (7)) react tothe
afford 1,2-cyclic,
phosphorylation
and if the 2,3-hydroxyl moieties are cis-disposed, then also 2,3-cyclic phosphates.
of ribose (7) proceeds initially like glyceraldehyde (GA), but then ‘with a twist’, the full activation Therefore, the
phosphorylation of ribose (7) proceeds initially like glyceraldehyde
potential of AmTP (the AmTP α-phosphate is activated twice, once with a pyrophosphate leaving(GA), but then ‘with a twist’, the
full activation potential of AmTP (the AmTP α-phosphate is activated twice, once with a
group, and once with an ammonia leaving group) is utilised to give ribose-1,2-cyclic phosphate
pyrophosphate leaving group, and once with an ammonia leaving group) is utilised to give ribose-
(ribo-20), and ribose-2,3-cyclic phosphate (21) (Scheme 7), which return the corresponding 2- or
1,2-cyclic phosphate (ribo-20), and ribose-2,3-cyclic phosphate (21) (Scheme 7), which return the
3-phosphates upon acidification. Conversely, due to trans-disposition of the 2,3-diol moiety of
corresponding 2- or 3-phosphates upon acidification. Conversely, due to trans-disposition of the 2,3-
arabinose
diol moiety(22),ofsimilar 3-phosphorylation
arabinose cannot lead to
(22), similar 3-phosphorylation the formation
cannot lead to the of a 2,3-cyclic
formation phosphate,
of a 2,3-cyclic
andphosphate,
therefore, andonlytherefore,
deliversonly arabinose-1,2-phosphate (arabino-20). Further ammonolysis
delivers arabinose-1,2-phosphate (arabino-20). Further ammonolysis of AmTP
affords diamidophosphate (DAP), which phosphorylates aldoses by
of AmTP affords diamidophosphate (DAP), which phosphorylates aldoses by a similar mechanisma similar mechanism to AmTP,
but to
does not require 2+
AmTP, but doesMg , and inMg
not require the ,case
2+ and inof the
aldopentose sugars, higher
case of aldopentose yields
sugars, higher are obtained
yields with DAP
are obtained
(71% vs. DAP
with 29%)(71%[67].vs.
The reaction
29%) of glycolaldehyde
[67]. The (GC) or glyceraldehyde
reaction of glycolaldehyde (GA) with
(GC) or glyceraldehyde (GA)DAPwithfurnishes
DAP
furnishes the α-phosphate
the α-phosphate derivatives inderivatives in excellent
excellent yield (>90%), yield
and(>90%),
we have and we have
recently recentlythis
exploited exploited thisand
reaction,
reaction,
these simpleand these
sugar simple sugar
phosphates, phosphates,
to develop to develop
a network a network of plausible
of prebiotically prebiotically plausible
reactions thatreactions
synthesise
that
all of thesynthesise all of of
intermediates thetriose
intermediates
glycolysisof[68]. triose glycolysis
For example,[68]. For example, glyceraldehyde-2-
glyceraldehyde-2-phosphate (GA2P)
obtained by DAP phosphorylation can be readily oxidised to afford glyceric acid 2-phosphate acid
phosphate (GA2P) obtained by DAP phosphorylation can be readily oxidised to afford glyceric (19-2p),
2-phosphatein(19-2p),
or dehydrated or dehydrated
pH 7 phosphate bufferin pH 7 phosphate
to afford buffer topyruvaldehyde
phosphoenol afford phosphoenol pyruvaldehyde
(23) (Scheme 8), which
(23) (Scheme 8), which can be oxidised to deliver phosphoenol pyruvate (PEP)—biology’s highest
can be oxidised to deliver phosphoenol pyruvate (PEP)—biology’s highest energy phosphate—in
energy phosphate—in excellent yield, following the trajectory of triose glycolysis observed in extant
excellent yield, following the trajectory of triose glycolysis observed in extant biology [68].
biology [68].

Scheme Prebiotic
8. 8.
Scheme Prebioticsynthesis
synthesisofofglyceric
glyceric acid
acid 2-phosphate (21-2p),phosphoserine
2-phosphate (21-2p), phosphoserine (Ser-3p)
(Ser-3p) andand
phosphoenol pyruvate (PEP) from the phosphorylation of glycolaldehyde (GC) and glyceraldehyde
phosphoenol pyruvate (PEP) from the phosphorylation of glycolaldehyde (GC) and glyceraldehyde
(GA) with
(GA) diamidophosphate
with diamidophosphate (DAP).
(DAP).

5. Nucleophilic
5. Nucleophilic Phosphorylation
Phosphorylation
SomeSome of the
of the problems
problems associatedwith
associated withthe
the electrophilic
electrophilic phosphorylation
phosphorylation ofof
hydroxyl moieties
hydroxyl moieties
can be overcome by tethering strategies, however, though these strategies have yielded excellent
can be overcome by tethering strategies, however, though these strategies have yielded excellent
syntheses of simple sugar phosphates and even phosphoenol pyruvate (PEP) in water, most major
syntheses of simple sugar phosphates and even phosphoenol pyruvate (PEP) in water, most major
Life 2017, 7, 31 10 of 23

phosphorylation targets in prebiotic chemistry are isolated hydroxyl moieties (e.g., nucleosides) which
do not have suitably positioned carbonyl moieties to facilitate delivery of phosphate. Electrophilic
phosphorylation strategies that employ substrate tethering to facilitate intramolecular delivery
of phosphate, can rely upon very weakly activated electrophilic phosphorus centres, however,
intermolecular delivery often requires higher levels of activation, making these strategies particularly
susceptible to competitive hydrolysis. Accordingly, it is valuable to consider other approaches.
The incredibly important role that phosphate ionization plays in the character of phosphates
(at physiological pH), due its low first and second pKa s (pKa = 2.2, 7.2, 12.3), has already been
noted [2], but this ionization also opens a different perspective on phosphate chemistry at neutral
pH. Whilst ionization provides kinetic stability to phosphate (di)esters, it also imparts significant
nucleophilicity to phosphate/phosphate monoesters at neutral pH, where the second ionization state is
readily accessed. This nucleophilicity is implicitly exploited in the electrophilic activation of phosphate,
for example, through reaction of phosphate with cyanate (4) to accrue electrophilic activation, however,
if electrophilic activation can be accrued in the substrate, then the direct phosphorylation of the
substrate can be achieved through reaction with phosphate in water. This is a particularly advantageous
strategy because phosphate (unlike hydroxyl groups) is significantly more nucleophilic than water
at neutral pH. This is a classic method to introduce ester and phosphoester moieties in organic
synthesis (e.g., Mitsonobu-type reactivity), however, at first glance, the hydroxyl motifs that are
key phosphorylation targets in prebiotic chemistry, do not easily lend themselves to activation as
electrophiles under prebiotic conditions. These problems are, however, a result of perspective, and
can be overcome by considering the approach to activation. Rather than specifically synthesising a
hydroxyl moiety and then pursuing subsequent activation of this hydroxyl group (following a classic
Mitsonobu strategy), which would be very challenging under prebiotic constraints, simple activating
strategies can be built into a chemical synthesis of a substrate from the start, without requiring an
intermediate hydroxyl moiety.
During a seminal contribution to the investigation of prebiotic sugar phosphate synthesis,
Eschenmoser and co-workers reported an exemplary nucleophilic phosphorylation in aqueous
solution using orthophosphate. They observed the phosphorylation of oxarinecarbonitrile 24 to
yield glycoaldehyde phosphate (GCP), or its cyanohydrin (GCP·HCN), in good yields under basic
conditions (Scheme 9) [69]. Similar chemistry was used by the same group in the synthesis of
phosphoserine (12), constitutionally related to glycoaldehyde phosphate (GCP), by opening of
aziridine-2-carbonitrile (25) in moderate yield (26, 50% after recrystallization) in acetonitrile [70].
The prebiotic synthesis of oxirane 24 or aziridine 25 has yet to be demonstrated, but these reactions
provide valuable mechanistic insights; they make use of the nucleophilicity of phosphate to achieve
substrate phosphorylation, rather than relying on addition to electrophilically activated phosphate.
Recently, we have exploited a similar approach to demonstrate a prebiotic synthesis of
aminooxazoline-5’-phosphates (27), which are key intermediates in the prebiotic synthesis of
nucleotides. Previously, the synthesis of 27 had been achieved by reaction of ribose-5’-phosphate
(R5P) and cyanamide (3) [71], or glyceraldehyde-3-phosphate (G3P) and 2-aminooxazole (2AO) [72].
However, neither R5P nor G3P are prebiotically accessible, moreover, under neutral and basic
conditions, G3P undergoes facile (E1cB) elimination to give pyruvaldehyde (28) [73,74]. To exploit
nucleophilic phosphorylation, but now in a prebiotically plausible system, we began our synthesis
with acrolein (29) (Scheme 10). Oxidation of 29 at near neutral pH by hydrogen peroxide furnishes
an excellent yield of glycidaldehyde (30) (>90% at pH 7.5–9). This reaction provides the oxirane
moiety and activation required for phosphorylation in water. Glycidaldehyde (30) can then be
directly phosphorylated in neutral aqueous solution by inorganic phosphate to afford G3P, but G3P,
as expected, rapidly eliminates. This elimination can be readily sidestepped, however, if glycidaldehyde
(30) first reacts with 2-aminooxazole (2AO), and then phosphate. The reaction of 2AO and 30 yield
a five-carbon sugar moiety that is activated (at the C5’-carbon atom) to nucleophilic substitution.
The epoxide moiety of intermediate 31 can be phosphorylated by direct addition of inorganic
Life 2017, 7, 31 11 of 23

phosphate to afford 27 in the first prebiotically plausible reaction process that specifically delivers
the Life
natural 5’-phosphorylation patterns observed in canonical nucleotides [75]. Reversing
2017, 7, 31 11 of 23
the
order of aminooxazoline assembly not only yielded a highly selective 5’-phosphorylation, but also,
by introducing the phosphate at
the order of aminooxazoline a distalnot
assembly position from the
only yielded anomeric
a highly centre,
selective prohibits the previous
5’-phosphorylation, but
also, by introducing
deleterious the phosphate
E1cB elimination. The useatofa nucleophilic,
distal positionrather
from the
thananomeric centre,phosphorylation,
electrophilic, prohibits the
previous
opens a wide deleterious
palette of siteE1cB elimination.
selective reactivity,The
anduse of warrants
clearly nucleophilic, rather
further than electrophilic,
investigation in the context
phosphorylation, opens a wide palette of site selective reactivity, and
of prebiotic chemistry, however nucleophilic phosphorylation strategies must be carefully clearly warrants considered
further
investigation
as reactivity in the context
is necessarily of prebiotic
orchestrated bychemistry,
the precedinghowever nucleophilic
chemical phosphorylation
reactions, and therefore,strategies
by in-built
must be carefully considered as
activation accrued within an organic substrate.reactivity is necessarily orchestrated by the preceding chemical
reactions, and therefore, by in-built activation accrued within an organic substrate.

Scheme
Scheme Nucleophilic
9. 9. Nucleophilicphosphorylation
phosphorylation in
in water. Oxirane(24)
water. Oxirane (24)and
andaziridine
aziridine (25)
(25) precursors
precursors of of
glycolaldehyde phosphate
glycolaldehyde phosphate (GCP)
(GCP)are phosphorylated
are phosphorylatedby
bynucleophilic
nucleophilic attack of inorganic
attack of inorganic phosphate
phosphate in
water or acetonitrile,
in water respectively.
or acetonitrile, respectively.

Scheme
Scheme 10.10. Synthesis
Synthesis of
of aminooxazoline
aminooxazoline5’-phosphate (27) (27)
5’-phosphate by nucleophilic phosphorylation.
by nucleophilic The
phosphorylation.
reaction of glycidaldehyde (30) and 2-aminooxazole (2AO) followed by phosphorylation
The reaction of glycidaldehyde (30) and 2-aminooxazole (2AO) followed by phosphorylation of 5’- of
activated oxirane 31 by inorganic phosphate yields aminooxazoline 5’-phosphate (27). Direct addition
5’-activated oxirane 31 by inorganic phosphate yields aminooxazoline 5’-phosphate (27). Direct addition
of phosphate to glycidaldehyde (30) affords G3P, which decomposes to methyl glyoxal (28).
of phosphate to glycidaldehyde (30) affords G3P, which decomposes to methyl glyoxal (28).
6. Phosphorylation in Dry-States Using Condensed Phosphates
As noted above, the competing effects of water have a tendency to significantly lower the yield
of phosphorylation reactions. Accordingly, excluding water through the simple process of
Life 2017, 7, 31 12 of 23

6. Phosphorylation in Dry-States Using Condensed Phosphates

As noted above, the competing effects of water have a tendency to significantly lower the yield of
phosphorylation reactions. Accordingly, excluding water through the simple process of evaporation
offers a remarkably interesting scenario in which to investigate prebiotic phosphorylation. Drying is
Life 2017, 7, 31 12 of 23
readily achieved by evaporation of water from a desired phosphorylation, and is a process that is very
easily envisaged offers
evaporation on the aearly Earth. These
remarkably dry-statescenario
interesting reactionsinarewhich of particular interest, because
to investigate prebioticthey
obfuscate the requirement
phosphorylation. Dryingfor condensing
is readily agents,achieved andbyrelyevaporation
only upon physical of water processing
from aof desired
phosphate
richphosphorylation,
media. and is a process that is very easily envisaged on the early Earth. These dry-state
Dry-state
reactions arephosphorylation
of particular interest,usingbecause
orthophosphate as thethe
they obfuscate source of phosphorus
requirement has been
for condensing applied
agents,
andsynthesis
to the rely only of upon physical
a wide range processing
of moleculesof phosphate rich media.
of prebiotic interest, however, phospholipids [76–78],
Dry-state[79–81],
and nucleotides phosphorylation
which both using orthophosphate
carry a phosphateasmoiety
the source of phosphorus
essential has been applied
to their function/structure,
to the synthesis of a wide range of molecules of prebiotic interest,
have received the most attention. The phosphorylation of nucleotides, for example, proceeds however, phospholipids [76–78],
rapidly
and nucleotides [79–81], which both carry a phosphate moiety ◦ essential
at high temperatures (16% phosphorylation after 2 h at 180 C) [79], but requires several months to to their function/structure,
have similar
achieve receivedresults
the most attention.
at milder The phosphorylation
temperatures of nucleotides,
(65–85 ◦ C). Total yieldsfor example,
are limitedproceeds rapidly of
by equilibration
at high temperatures (16% phosphorylation after 2 hr at 180 °C) [79], but requires several months to
phosphorylated and non-phosphorylated products [81], but high levels of phosphate incorporation
achieve similar results at milder temperatures (65–85 °C). Total yields are limited by equilibration of
can be driven by the irreversible formation of 2’,3’-cyclic phosphate. The efficiency of dry-state
phosphorylated and non-phosphorylated products [81], but high levels of phosphate incorporation
phosphorylation reactions is significantly improved by the inclusion of urea (8) [82]. After evaporation,
can be driven by the irreversible formation of 2’,3’-cyclic phosphate. The efficiency of dry-state
ureaphosphorylation
(8) acts as both areactionscatalyst is forsignificantly
phosphoryl improved
transfer and by asthea pseudo-solvent,
inclusion of urea providing
(8) [82].fluidity
After at
elevated temperatures. Phosphorylation of nucleosides with mixtures
evaporation, urea (8) acts as both a catalyst for phosphoryl transfer and as a pseudo-solvent, of urea (8), ammonium chloride
andproviding
various phosphates (Na2 HPO ◦
fluidity at elevated 4 and Ca5Phosphorylation
temperatures. (PO4 )3 OH) at temperatures
of nucleosides ranging fromof60urea
with mixtures to 100
(8), C,
afford nucleotides.
ammonium Highand
chloride degrees
various of phosphates
total phosphate incorporation
(Na2HPO 4 and Ca5(PO are
4)3observed (>96% for pyrimidine
OH) at temperatures ranging
nucleotides).
from 60 toAmmonium
100 °C, afford salts are also commonly
nucleotides. High degrees usedoftototal
avoid carbamylation
phosphate (upon loss
incorporation of ammonia
are observed
from (>96%
ureafor pyrimidine
(8)) and assistnucleotides). Ammoniumofsalts
with the acidification theare also commonly
reaction mixture,used which to avoid
promotescarbamylation
phosphorus
(upon[83–86].
transfer loss of ammonia from urea (8)) and assist with the acidification of the reaction mixture, which
promotes
The precise phosphorus
mechanistictransferrole[83–86].
of urea (8) has not been proven, however, it is likely that urea (8)
displaces water from a tautomericof form
The precise mechanistic role urea (8)of has not been proven,
monoanionic however,
phosphate, in itwhich
is likelythethat urea (8)
charge state
displaces water from a tautomeric form of monoanionic phosphate,
facilitates dissociative loss of water, and nucleophilic attack (by urea (8)), to generate an activated in which the charge state
facilitates dissociative loss of water, and nucleophilic attack (by urea (8)), to generate an activated
ureidophosphate intermediate 32 capable of transferring phosphate between hydroxyl moieties
ureidophosphate intermediate 32 capable of transferring phosphate between hydroxyl moieties
(Scheme 11). Further evidence for this mechanism is found in the reversible phosphorylation of
(Scheme 11). Further evidence for this mechanism is found in the reversible phosphorylation of
hydroxyl groups, but the irreversible synthesis of 2’,3’-cyclic phosphates. Accordingly, and necessarily,
hydroxyl groups, but the irreversible synthesis of 2’,3’-cyclic phosphates. Accordingly, and
a different mechanism
necessarily, a different formechanism
cyclisationfor ofcyclisation
2’,3’-cyclicofphosphates must operate
2’,3’-cyclic phosphates mustunder
operatethese conditions.
under these
It seems
conditions. It seems likely that due to the high effective molarity of the 2’-hydroxyl to a 3’-phosphatevice
likely that due to the high effective molarity of the 2’-hydroxyl to a 3’-phosphate (or versa),
(or vice
the versa),
phosphorylation mechanism
the phosphorylation can switch
mechanism to antoassociative
can switch an associative mechanism,
mechanism, therefore
therefore allowing
allowing cyclic
phosphate synthesis, but
cyclic phosphate prohibiting
synthesis, the degradation
but prohibiting of 2’,3’-cyclic of
the degradation phosphates
2’,3’-cyclicunder these conditions.
phosphates under
It isthese
the mechanistic
conditions. Itswitch that is responsible
is the mechanistic switch that foristhe highly effective
responsible accumulation
for the highly of phosphate in
effective accumulation
of phosphate in ribonucleotides under these conditions, and
ribonucleotides under these conditions, and therefore potentially an important contributing therefore potentially an important
factor in
the contributing factor in the selection
selection of ribonucleotides as keyof ribonucleotides
metabolites to as key metabolites
support life fromto support chemistry.
prebiotic life from prebiotic
chemistry.

Scheme Proposed
11.11.
Scheme Proposedmechanism
mechanismfor
forurea-mediated dissociativedry-state
urea-mediated dissociative dry-state phosphorylation.
phosphorylation.

This dry-state urea-mediated phosphorylation protocol has been applied to the phosphorylation
This dry-state urea-mediated phosphorylation protocol has been applied to the phosphorylation
of various anhydronucleotides to good effect [86,87]. Upon phosphorylation of anhydronucleotides,
of various anhydronucleotides to good effect [86,87]. Upon phosphorylation of anhydronucleotides,
under the conditions of urea-mediated phosphoryl-transfer, a third mechanism comes into play,
under the conditions of urea-mediated phosphoryl-transfer, a third mechanism comes into play,
exploiting the activation inherent in the anhydronucleotide bond to synthesise a cyclic phosphate;
exploiting the than
but rather activation inherent in
by dissociative the anhydronucleotide
phosphoryl-transfer, bond by
is followed to synthesise
associative acyclisation,
cyclic phosphate;
now
but dissociative
rather thanphosphoryl-transfer
by dissociative phosphoryl-transfer, is followed by associative
is followed by intramolecular rearrangement by attack cyclisation,
of (tethered)now
nucleophilic phosphate on an activated carbon atom. Akin to the activation that was discussed above,
built into the synthesis of an oxirane, here, the stepwise assembly of an anhydronucleotide can be
used to chemospecifically direct the activation of a carbon atom to facilitate the synthesis of a new
phosphate group. As this phosphorylation mechanism exploits the activation of the C2’-carbon atom
Life 2017, 7, 31 13 of 23

dissociative phosphoryl-transfer is followed by intramolecular rearrangement by attack of (tethered)

nucleophilic phosphate on an activated carbon atom. Akin to the activation that was discussed
above, built into the synthesis of an oxirane, here, the stepwise assembly of an anhydronucleotide
can be used to chemospecifically direct the activation of a carbon atom to facilitate the synthesis
of a new phosphate group. As this phosphorylation mechanism exploits the activation of the
C2’-carbon atom to SN 2-nucleophilic displacement, anhydro-arabinofuranosyl nucleosides arabino-33
Life 2017, 7, 31 13 of 23
and arabino-34 are phosphorylated, and rearrange to deliver the natural ribofuranosyl nucleotides as
their
to Sstable 2’,3’-cyclic
N2-nucleophilic phosphatesanhydro-arabinofuranosyl
displacement, 35 and 36. The phosphorylation nucleosides of anhydronucleotides
arabino-33 and arabino-34 displays
remarkable selectivity
are phosphorylated, andforrearrange
phosphorylation
to deliver theof the 3’-OH
natural [86,87], asnucleotides
ribofuranosyl a result ofasboth their the kinetic
stable
→π* donation
(n2’,3’-cyclic supresses
phosphates nucleophilicity
35 and of the 5’-OH)
36. The phosphorylation of and thermodynamicdisplays
anhydronucleotides (irreversible 2’,3’-cyclic
remarkable
selectivity synthesis)
phosphate for phosphorylation of the [86–88].
characteristics 3’-OH [86,87], as a result
To further of boththe
emphasise thepotential
kinetic (n→π*
of thisdonation
strategy, we
supresses
have recently nucleophilicity of the 5'-OH) and
applied the urea-mediated thermodynamic
phosphorylation to (irreversible
the divergent 2’,3’-cyclic
synthesisphosphate
of pyrimidine
synthesis)
and characteristics
8-oxo-purine [86–88]. To wherein
ribonucleotide, further emphasise the potential
it is remarkable to of thisthat
note strategy, we have
a minor recently
modification of
applied the urea-mediated phosphorylation to the divergent synthesis
the purine moiety (C8 oxidation) leads to divergent reaction pathways to canonical pyrimidine of pyrimidine and 8-oxo-
purine ribonucleotide,
nucleotides whereinribonucleotides,
and 8-oxo-purine it is remarkable tofrom note that
one acommon
minor modification
intermediate. of the purine moiety
Importantly, though
(C8 oxidation) leads to divergent reaction pathways to canonical pyrimidine nucleotides and 8-oxo-
the specific heterocyclic motifs of the intermediate anhydro-nucleotides arabino-33 and arabino-34
purine ribonucleotides, from one common intermediate. Importantly, though the specific
(Scheme 12) are different, they both display n→π* suppression of 5’-OH nucleophilicity, and both
heterocyclic motifs of the intermediate anhydro-nucleotides arabino-33 and arabino-34 (Scheme 12) are
undergo 3’-OH selective phosphorylation, followed by intramolecular inversion to yield the natural
different, they both display n→π* suppression of 5’-OH nucleophilicity, and both undergo 3’-OH
β-ribo-stereochemistry.
selective phosphorylation, followed by intramolecular inversion to yield the natural β-ribo-
Once again, through careful consideration of the method used to construct the organic substrates,
stereochemistry.
it wasOnce possibleagain,tothrough
build two careful substrates (one of
consideration purine and oneused
the method pyrimidine)
to construct that
theboth directed
organic
phosphorylation
substrates, it waschemistry
possible tovia the two
build samesubstrates
chemical(one strategy.
purine Both
andpathways exploitthat
one pyrimidine) dry-state dissociative
both directed
phosphoryl-transfer,
phosphorylation chemistry that is (partially)
via the same controlled
chemicalthrough
strategy.stereoelectronic
Both pathwayseffects, exploitboth pathways
dry-state
dissociative phosphoryl-transfer, that is (partially) controlled through
exploit tethered nucleophilic attack to deliver a new cyclic phosphate moiety, and finally, stereoelectronic effects, both both
pathways exploited
pathways exploit tethered
in-builtnucleophilic
electrophilicattackactivation
to deliver aofnew cyclic phosphate
a carbon atom through moiety,theandformation
finally, of
anboth pathways exploited
anhydronucleotide in-built
bond to theelectrophilic
C2’ carbon activation
atom. ofThough
a carbonmoreatomwork
through the formation
is required of
to realise a
an anhydronucleotide bond to the C2’ carbon atom. Though more work is
prebiotic synthesis of the canonical purines, the continued success of these strategies warrants further required to realise a
prebiotic synthesis
investigation of the canonical
of dry-state nucleotide purines, the continued success of these strategies warrants further
phosphorylations.
investigation of dry-state nucleotide phosphorylations.
CN NH2
O O O CN
N NH3 N H2N CN N
HO HO HO
NH2 SR N
HO O HO O HO O
arabino-41

CN (1) HC(NH)NH2

N N
O O Base O
N H2PO4- H2PO4- N NH2
HO HO HO
NH2 CO(NH2)2 N
N CO(NH2)2
HO O O O HO O
(8) P (8)
arabino-33 arabino-34
O O
35 Base = cytosine
36 Base = 8-oxo adenosine
Scheme 12. Divergent synthesis of pyrimidine ribonucleotide 35 and 8-oxo-purine ribonucleotide 36
Scheme 12. Divergent
under prebiotic synthesis of pyrimidine ribonucleotide 35 and 8-oxo-purine ribonucleotide 36
conditions.
under prebiotic conditions.
Beyond the phosphorylation of nucleotides, the reversible characteristics of urea-mediated
phosphorylations have also been exploited
Beyond the phosphorylation to furnish the
of nucleotides, phospholipid-type molecules selectively
reversible characteristics from
of urea-mediated
a mixture of alcohols. Dry-state phosphorylation exploits evaporation to remove water
phosphorylations have also been exploited to furnish phospholipid-type molecules selectively from
phosphorylation reactions, and accordingly, simple alcohols (such as the oxygenate products of
from a mixture of alcohols. Dry-state phosphorylation exploits evaporation to remove water from
Fischer–Tropsch synthesis) can be readily fractionated by evaporation during these phosphorylation
phosphorylation reactions, and accordingly, simple alcohols (such as the oxygenate products of
reactions. Long chain alcohols, such as decanol (37), can be phosphorylated in preference to shorter
Fischer–Tropsch synthesis) can be readily fractionated by evaporation during these phosphorylation
chain alcohols, such as hexanol (38) and ethanol (39), to selectively yield decyl-phosphate (40), by
simple virtue of comparative volatility (Scheme 13) [29].
Life 2017, 7, 31 14 of 23

reactions. Long chain alcohols, such as decanol (37), can be phosphorylated in preference to shorter
chain alcohols, such as hexanol (38) and ethanol (39), to selectively yield decyl-phosphate (40),
by simple
Life 2017,virtue
7, 31 of comparative volatility (Scheme 13) [29]. 14 of 23

Scheme Phosphorylation
13. 13.
Scheme Phosphorylationof long chains
of long alcohols
chains such
alcohols as decanol
such (37)(37)
as decanol is selective over
is selective shorter
over chain
shorter
alcohols
chain due to the
alcohols different
due volatilities
to the different of theseofcompounds.
volatilities these compounds.

The ability of urea (8) to facilitate phosphoryl-transfer derives from the nucleophilicity of the urea
The ability of urea (8) to facilitate phosphoryl-transfer derives from the nucleophilicity of
oxygen atom towards phosphorus electrophiles. This property is certainly not unique to urea (8), for
the urea oxygen atom towards phosphorus electrophiles. This property is certainly not unique to
example, formamide will also (albeit less efficiently, likely due to significantly reduced nucleophilicity
urea (8), for example, formamide will also (albeit less efficiently, likely due to significantly reduced
of the amide vs. urea oxygen) promote phosphoryl-transfer [86,87,89–92].
nucleophilicity of theefficiency
The excellent amide vs.andurea oxygen) and
selectivity, promote phosphoryl-transfer
the remarkable simplicity of[86,87,89–92].
amide/urea-mediated
The excellent efficiency
phosphorylation, and selectivity,
will no doubt and the
see its continued remarkable
application simplicity
in unpicking theoforigins
amide/urea-mediated
of life, and the
phosphorylation,
sheer volume of research, and the number of years that these phosphorylation strategiesofhave
will no doubt see its continued application in unpicking the origins life, been
and the
sheer
employed, are a testament to the efficiency and reproducibility of dry-state phosphorylation as been
volume of research, and the number of years that these phosphorylation strategies have a
employed,
method, are
anda perhaps
testament to more
this, the efficiency and reproducibility
than anything of dry-state
else, is an indication that it phosphorylation
might come to beas a
recognised
method, as a keythis,
and perhaps element
moreinthan
facilitating
anythingtheelse,
incorporation of phosphate
is an indication in living
that it might systems.
come to be recognised
as a key element in facilitating the incorporation of phosphate in living systems.
7. Phosphate as a Catalyst
7. Phosphate
As hasasbeena Catalyst
made clear in the above discussions, phosphate is an important constitutional
component
As has been of life,
made andclear
numerous
in themethods for its incorporation
above discussions, phosphate intoismetabolites
an important under prebiotic
constitutional
conditions are being developed. Another feature of prebiotic phosphate
component of life, and numerous methods for its incorporation into metabolites under prebiotic chemistry, which has
received far less attention, but can be nonetheless striking in its application,
conditions are being developed. Another feature of prebiotic phosphate chemistry, which has received are the catalytic and
(pH/chemical) buffering capabilities of phosphate. Whilst phosphate is inherently required as a
far less attention, but can be nonetheless striking in its application, are the catalytic and (pH/chemical)
constitutional component for the assembly of RNA, phosphoenol pyruvate (PEP), phospholipids,
buffering capabilities of phosphate. Whilst phosphate is inherently required as a constitutional
etc., it has also been shown to have a critical, multi-faceted, and often very subtle non-structural role
component for the assembly of RNA, phosphoenol pyruvate (PEP), phospholipids, etc., it has also
in prebiotic synthesis. For example, during the prebiotic synthesis of pyrimidine ribonucleotides
been shownby
reported to Powner
have a critical,
et al. [87],multi-faceted,
all steps in theand often from
sequence very 2-subtle non-structural
and 3-carbon role in blocks
atom building prebiotic
synthesis. Forout
are carried example, during of
in the presence thephosphate
prebiotic(Scheme
synthesis 14).ofThe
pyrimidine
phosphateribonucleotides
is ultimately the reported
source of by
Powner
phosphorus for urea-mediated phosphorylation and 2′-inversion [87], and it was therefore are
et al. [87], all steps in the sequence from 2- and 3-carbon atom building blocks carried
deemed
out important
in the presence of phosphate
to demonstrate (Scheme
phosphate 14).beThe
could phosphate
present from theis ultimately the source the
start, and throughout of phosphorus
reaction
for urea-mediated
sequence, however, phosphorylation
phosphate alsoand 20 -inversion
performs numerous[87],other
and itcrucial
was therefore deemed
roles at critical important
stages in the to
synthesis.phosphate
demonstrate For example, whilst
could most steps
be present from proceed efficiently
the start, at near-neutral
and throughout pH, thesequence,
the reaction formationhowever,
of 2-
aminooxazole (2AO), from glycolaldehyde (GC) and cyanamide (3), initially
phosphate also performs numerous other crucial roles at critical stages in the synthesis. For example, appeared to require
alkaline
whilst mostconditions,
steps proceed to allow the keyatbase-catalysed
efficiently near-neutral stepspH, theto proceed
formation smoothly [93]; indeed, a(2AO),
of 2-aminooxazole complex from
mixture of intermediate
glycolaldehyde (GC) and addition
cyanamide products were returned
(3), initially appeared in neutral water.alkaline
to require The requirement
conditions, for high-
to allow
the pH
keyconditions in thissteps
base-catalysed first step are negated,
to proceed however,
smoothly [93];inindeed,
the presence of phosphate
a complex mixturebuffer. A high
of intermediate
yield of 2AO is obtained at pH 7 in phosphate solution, due to the efficient general acid–base catalysis
addition products were returned in neutral water. The requirement for high-pH conditions in this first
exhibited by the phosphate anion/dianion (pKa 7.2). By lowering the pH of this first step (through
step are negated, however, in the presence of phosphate buffer. A high yield of 2AO is obtained
general acid–base catalysis), phosphate brings this step into consonance with the rest of the route that
at pH 7 in phosphate solution, due to the efficient general acid–base catalysis exhibited by the
requires near neutral pH conditions (pH 6–7). Furthermore, excess cyanamide (3), which has the
phosphate
potentialanion/dianion (pKastages,
to interfere in later 7.2). By lowering upon
particularly the pH of this firstofstep
introduction (through general
glyceraldehyde (GA), acid–base
slowly
catalysis),
reacts in the presence of phosphate to afford urea (8), a compound that is later instrumental innear
phosphate brings this step into consonance with the rest of the route that requires
catalysing the phosphorylation/rearrangement step (discussed above). Phosphate also acts as a pH
buffer in the reaction of 2AO with glyceraldehyde (GA) to give the aminooxazolines 41, as a mixture
 to get to the natural β-cytidine (β-42), which has a greatly adverse impact on the overall yield, but
perhaps more importantly (without pH buffering), poor regioselectivity of cynanovinylation is
observed, and a wide range of by-products result alongside α-42 and 43, from these reactions.
However, if the cyanovinylation is performed in the presence of phosphate buffer at pH 6.5,
anhydronucleoside (33) is furnished in near quantitative yield. The anhydronucleotide linkage of 33
is unstable
Life 2017, 7, 31 without phosphate, due to the tendency for the solution pH to increase as the reaction 15 of 23
proceeds, but hydrolysis to α-42 or 43 is completely suppressed in the presence of phosphate buffer.
The arabino-anhydrocytidine (arabino-33) can then be efficiently converted to β-cytidine-2’,3’-cyclic
neutral pH conditions
phosphate (pHurea-mediated
(35) through 6–7). Furthermore, excess cyanamide
phosphorylation (3), which hasinversion.
and C2’-stereochemical the potential
The to
interfere
overall efficiency of the cyanovinylation reaction is markedly improved in the presence of phosphate, in
in later stages, particularly upon introduction of glyceraldehyde (GA), slowly reacts
the bypresence of the
stabilising phosphate to afford
solution pH urea
(alkaline), (8), a compound
hydrolysis that is later instrumental
of both anhydronucleotide in catalysing
33 and cyanoacetylene
the (1)
phosphorylation/rearrangement
are minimised, which would otherwise step (discussed above).
limit the yield. Phosphate
Finally, also acts
the phosphate in as
thisa reaction
pH buffer
actsreaction
in the as a chemical
of 2AO buffer;
withexcess cyanoacetylene
glyceraldehyde (GA)(1)to
is sequestered by phosphate to41,
give the aminooxazolines giveascyanovinyl
a mixture of
phosphate (44),
diasteriomers preventing overreaction
(arabino:ribo:xylo:lyxo with the
15:25:6:4). desired
Ribose nucleotide product,
aminoxazoline and,isimportantly,
(ribo-41) 44 is of
the least soluble
a phosphorylating agent in its own right that goes on to react further with phosphate,
these, and can be isolated by direct crystallization; as such, the arabinose aminooxazoline (arabino-41) to yield
pyrophosphate
becomes (PPi). Pyrophosphate
the major product can be
in the supernatant [94].subsequently
The precipitatedusedribose
in the anhydronucleotide
aminooxazoline (ribo-41)
phosphorylation step to afford improved yield, and reduce hydrolytic by-products. Thus, phosphate
can be redissolved in phosphate solution, and equilibrated with the arabino-configuration via a
behaves not just as a reagent in prebiotic nucleotide synthesis, but also as a catalyst, a pH buffer, and
phosphate-mediated isomerisation process, where again, phosphate acts as a general acid–base catalyst
a chemical buffer, removing potentially detrimental reactive species such as cyanoacetylene (1) and
for the epimerisation of C2’-carbon atom of the aminooxazolines 41 [95].
cyanamide (3) from solution.
O
Pi
OH O O
O H 2NCN (3) N N N
OH GA HO HO
OH NH 2 + NH 2
O NH 2 O O
Pi Pi HO HO
GC 2AO arabino-41 ribo-41

O
O O
P
O CN
Pi N O Pi (1) No Pi
HO
NH 2 +
CO(NH 2) 2 O N
HO CN
(8)
arabino-33 44

O O O
N N N
NH 2 NH 2 NH 2
O N O N O N
HO HO + HO

O O HO OH HO OH
P
O O α-42 43
35

Scheme Prebiotic
14. 14.
Scheme synthesis
Prebiotic synthesisofofpyrimidine
pyrimidine ribonucleotides (35),ininwhich
ribonucleotides (35), which each
each step
step is facilitated
is facilitated by by
phosphate (Pi).
phosphate (Pi).

Phosphate plays a particularly essential role in the reaction of the aminooxazolines (41)
with cyanoacetylene (1). In water, the reaction of 41 and 1 results in the formation of cytidine
nucleotides [71,96]; ribose aminooxazoline (ribo-41) affords α-cytidine (α-42), and arabinose
aminooxazoline (arabino-41) gives β-arabinosyl-cytidine (43). These products, α-42 and 43, require
further stereochemical manipulation to get to the natural β-cytidine (β-42), which has a greatly adverse
impact on the overall yield, but perhaps more importantly (without pH buffering), poor regioselectivity
of cynanovinylation is observed, and a wide range of by-products result alongside α-42 and 43, from
these reactions. However, if the cyanovinylation is performed in the presence of phosphate buffer
at pH 6.5, anhydronucleoside (33) is furnished in near quantitative yield. The anhydronucleotide
linkage of 33 is unstable without phosphate, due to the tendency for the solution pH to increase
as the reaction proceeds, but hydrolysis to α-42 or 43 is completely suppressed in the presence
of phosphate buffer. The arabino-anhydrocytidine (arabino-33) can then be efficiently converted to
β-cytidine-2’,3’-cyclic phosphate (35) through urea-mediated phosphorylation and C2’-stereochemical
inversion. The overall efficiency of the cyanovinylation reaction is markedly improved in the presence
of phosphate, by stabilising the solution pH (alkaline), hydrolysis of both anhydronucleotide 33 and
cyanoacetylene (1) are minimised, which would otherwise limit the yield. Finally, the phosphate
in this reaction acts as a chemical buffer; excess cyanoacetylene (1) is sequestered by phosphate to
give cyanovinyl phosphate (44), preventing overreaction with the desired nucleotide product, and,
Life 2017, 7, 31 16 of 23

importantly, 44 is a phosphorylating agent in its own right that goes on to react further with phosphate,
to yield pyrophosphate (PPi). Pyrophosphate can be subsequently used in the anhydronucleotide
phosphorylation step to afford improved yield, and reduce hydrolytic by-products. Thus, phosphate
behaves not just as a reagent in prebiotic nucleotide synthesis, but also as a catalyst, a pH buffer, and
a chemical buffer, removing potentially detrimental reactive species such as cyanoacetylene (1) and
cyanamide (3) from solution.
The isomerisation of glyceraldehyde (GA) to dihydroxyacetone (DHA) is promoted by phosphate
catalysis. This equilibrium, which strongly favours DHA, had been seen as an undesired reaction,
with the potential to hamper the formation of aminooxazolines (41) en route to ribonucleotides. Though
photoreduction of DHA affords a lipid precursor (glycerol) and valine precursor (acetone) [89], DHA
would be detrimental to selective ribonucleotide synthesis, opening an interesting conundrum for
the origins of RNA. The phosphate rich conditions that favour nucleotide assembly, catalyse the loss
of GA (an essential nucleotide precursor) to DHA. In pursuit of a solution to this ‘DHA-problem’,
we observed that incubation of DHA with β-mercaptanoacetaldehyde (BMA) and cyanamide (3)
in phosphate buffer (pH 7), produced pure glyceraldehyde aminal (45) in excellent yield (>85%)
(Scheme 15). Importantly, 45 is sequestered as a crystalline precipitate, and this crystallisation inverts
the thermodynamic preference for DHA > GA [97,98]. In the same way, glycolaldehyde (GC) and
2-aminothiazole (2AT) produce the corresponding crystalline aminal (46), however, as glycolaldehyde
(GC) is symmetric with respect to Lobry de Bruyn–van Ekenstein transformation, and it cannot
equilibrate with a ketose isomer, therefore, crystallisation of 46 is significantly more rapid than
crystallisation of 45 from DHA. The precipitation of 45 is effectively time-resolved from precipitation
of 46 through phosphate catalysed Lobry de Bruyn–van Ekenstein rearrangement, because precipitation
of 45 requires slow release of GA from a more stable ketose isomer, DHA. Therefore, when complex
mixtures of aldoses and ketoses (up to 26 different sugars), and 2AT (or stoichiometric BMA and
cyanamide (3)) are incubated in phosphate solution, selective sequestration of firstly, glycolaldehyde
(GC), and then GA, occurs. This sequestration and resolution of C2 and then C3 sugars, solely
as their aldose forms, from complex mixtures, predicates the order that is required for selective
assembly of ribonucleotides from complex prebiotically plausible mixtures. Once again, phosphate
acts as an essential general acid–base catalyst in this sequestration and selection process, and the
phosphate-facilitated isomerization of DHA/GA is essential to the separation of these two important
aldoses for the synthesis of ribonucleotides [98].
As well as catalysing aldo–keto isomerisation (Lobry de Bruyn–van Ekenstein transformation),
phosphate has been shown to promote imine–enamine–aldehyde tautomerization, and the Amadori
rearrangement. For example, glycolaldehyde (GC) can be coupled to the purine precursor
5-amino-imidazole-4-carboxamide (47), via an amine linkage through Amadori rearrangement by
mild heating (60 ◦ C) in phosphate solution to furnish azepinomycin (48)—a guanine deaminase
inhibitor—in one protecting-group-free step, from low cost commercial materials (Scheme 16). This
Amadori rearrangement raises the prospect of an alternative, mild method for coupling sugars
with nitrogenous bases [99], but also highlights an important role that efficient prebiotic chemistry
can play in delivering synthetically valuable reaction strategies that can find general application in
organic chemistry. Recently, Szostak and co-workers have demonstrated the remarkable efficiency of
NMP activation by 2-aminoimidazole (2AI) [100], which is of particular note, due to its generational
relationship with 2-aminoxazole (2AO) and 2-aminothiazole (2AT) (that are discussed above) through
a phosphate catalysed Amadori rearrangement (Scheme 16) [98,101].
Recently, the prebiotic synthesis of high-energy (glycolysis) metabolite phosphoenol pyruvate
(PEP) was demonstrated [68], and once again, the synthesis highlighted the multiple roles that
phosphorus can occupy in robust prebiotic synthesis. The α-phosphorylation of glycolaldehyde
(GC) and glyceraldehyde (GA) by diamidophosphate (DAP) was rapidly and efficiently achieved in
phosphate solution, where phosphate’s buffering capacity acts to off-set the detrimental pH increase
due to release of ammonia, which had otherwise been found to inhibit phosphorylation, and required
Life 2017, 7, 31 17 of 23

manual and continuous pH adjustments to offset this effect. Incubation of glyceraldehyde-2-phosphate

(GA2P) in neutral phosphate solution then resulted in a high-yielding dehydration to furnish
phosphoenol pyruvaldehyde (23), the direct precursor (via oxidation) to PEP (Scheme 8). Similarly,
phosphate catalysis was demonstrated to efficiently convert glycolaldehyde phosphate (GCP) directly
to phosphoenol pyruvaldehyde (23), through a one-pot aldol condensation and elimination with
formaldehyde (49). Formation of glyceraldehyde phosphate (GA2P), from the aldol reaction of GCP
and 49, was known to occur slowly under alkaline conditions [63], but incubation at 60 ◦ C in phosphate
Life 2017, 7,
solution 31
allowed a phosphate catalysed aldolisation and subsequent dehydration to be carried 17 ofout
23

rapidly and cleanly at neutral pH, once again demonstrating the remarkable effect of phosphate
to be carried out rapidly and cleanly at neutral pH, once again demonstrating the remarkable effect
catalysis. The efficiency of phosphate catalysis, and the effective role that it can play catalysing proton
of phosphate catalysis. The efficiency of phosphate catalysis, and the effective role that it can play
transfer at near neutral pH in aqueous solution, suggests that there is a great deal more to discover
catalysing proton transfer at near neutral pH in aqueous solution, suggests that there is a great deal
with respect to the subtle effects that phosphate can induce within the context of prebiotic chemistry.
more to discover with respect to the subtle effects that phosphate can induce within the context of
One is struck by the simplicity of this catalyst and this simplicity, as well as its ideal pKa match for
prebiotic chemistry. One is struck by the simplicity of this catalyst and this simplicity, as well as its
neutral pH reactivity, which is likely, in part, responsible for its broad scope of application. New roles
ideal pKa match for neutral pH reactivity, which is likely, in part, responsible for its broad scope of
for phosphates will continue to be found, such as the remarkable hydrotropic properties of ATP that
application. New roles for phosphates will continue to be found, such as the remarkable hydrotropic
lead to enhanced protein solubility [102], and it is the combination and breadth of the various attributes
properties of ATP that lead to enhanced protein solubility [102], and it is the combination and breadth
of phosphate chemistry that define the value of phosphates in biology and biochemistry, but when
of the various attributes of phosphate chemistry that define the value of phosphates in biology and
these remarkable properties are considered with the deep-seated evolutionary history of phosphate in
biochemistry, but when these remarkable properties are considered with the deep-seated
life, it becomes history
evolutionary more pressing than ever
of phosphate in to respond
life, to themore
it becomes availability
pressingof phosphates
than ever toinrespond
a positive
tolight,
the
rather than retreating from this challenge.
availability of phosphates in a positive light, rather than retreating from this challenge.
O O
O OH
OH + SH +
OH
GC BMA GA

H2NCN (3)
HPO42-

O N O
OH + NH2 + HO OH
S
GC 2AT DHA

HPO42-

S S

N HN N N HN N

S OH S
N N OH
H H
OH
46 45
ppt after 2h ppt after 2-20d
(quant.) (>85%)

H2NCN (3) N
41
O NH2
2AO

Scheme 15. Resolution and sequestration of glycolaldehyde (GC) and glyceraldehyde (GA) from
Scheme 15. Resolution and sequestration of glycolaldehyde (GC) and glyceraldehyde (GA) from
aqueous solution. Phosphate acts as a general acid–base catalyst for the Lobry de Bruyn–van Ekenstein
aqueous solution. Phosphate acts as a general acid–base catalyst for the Lobry de Bruyn–van
transformation during the prebiotic selection of aldose sugars from complex sugar mixtures, facilitating
Ekenstein transformation during the prebiotic selection of aldose sugars from complex sugar
the resolution of aminals 45 and 46 en route to pentose aminooxazolines (41) and ribonucleotides.
mixtures, facilitating the resolution of aminals 45 and 46 en route to pentose aminooxazolines (41)
ppt = precipitate.
and ribonucleotides. ppt = precipitate.
Life 2017, 7, 31 18 of 23
Life 2017, 7, 31 18 of 23

a) O O
N NH 2 N NH
H 2PO 4-
+ GC OH
N H 2O N
H NH 2 H N
H
47 48

b)
O NH 2
NH 3 + GC
H 2PO 4-
N NH
+ H 2NCN (3) H 2O NH 2
2AI

Scheme
Scheme Phosphate
16.16. Phosphatecatalysed
catalysedAmadori
Amadori rearrangements. (a)Synthesis
rearrangements. (a) Synthesisofofazepinomycin
azepinomycin (48)
(48) byby
a a
one-pot phosphate
one-pot phosphatecatalysed
catalysedprotocol.
protocol.(b)
(b)Prebiotic
Prebiotic synthesis of 2-aminoimidazole
synthesis of 2-aminoimidazole(2AI).
(2AI).

8. Conclusions
8. Conclusions andOutlook
and Outlook
Significant
Significant progress
progress has been
has been made made
towards towards understanding
understanding the prebiotic
the prebiotic chemistrychemistry
of phosphorus, of
phosphorus, but is clear that the ‘phosphate problem’ is still to be conclusively
but is clear that the ‘phosphate problem’ is still to be conclusively resolved. However, efforts to resolved. However,
efforts to demonstrate further prebiotic chemistry of phosphates should not be hindered by this lack
demonstrate further prebiotic chemistry of phosphates should not be hindered by this lack of
of geophysical certainty; modern biochemical evidence suggests that phosphates are very likely to
geophysical certainty; modern biochemical evidence suggests that phosphates are very likely to have
have played a significant part in the chemical origins of life. Moreover, there are multiple potential
played a significant part in the chemical origins of life. Moreover, there are multiple potential solutions
solutions to the problem of phosphate availability, and several plausible phosphate-producing
to the problem of phosphate availability, and several plausible phosphate-producing environments
environments can now be considered. Volcanic activity may provide ortho-, pyro-, and
cantripolyphosphate
now be considered. Volcanic activity may provide ortho-, pyro-, and tripolyphosphate from
from apatite and basalt [38]; the same compounds can be obtained by the oxidation
apatite
of schreibersite [20–23], asame
and basalt [38]; the mineralcompounds can be obtained
found in meteorites. by the oxidation
Furthermore, instead ofoffocusing
schreibersite [20–23],
on ‘global’
a mineral found in meteorites. Furthermore, instead of focusing on ‘global’
phosphorus availability (in oceans), the essential role that phosphates play in biology may suggest phosphorus availability
(inthat
oceans), the essential
accumulation in localrole that phosphates
environments play inlakes,
(e.g., ponds, biology may suggest
or craters) playedthat accumulation
an essential role in in
local
defining the setting for prebiotic (phosphate) chemistry. In such environments, systems tailored tofor
environments (e.g., ponds, lakes, or craters) played an essential role in defining the setting
prebiotic
phosphate (phosphate)
release and chemistry. In such
accumulation, suchenvironments,
as cation/anion systems tailoredcan
fractionation, to be
phosphate release and
readily envisaged.
accumulation,
If these ponds, such
lakes,asor cation/anion
pools are linked fractionation, can beorreadily
together by rivers stream envisaged.
systems, and/or If these ponds, lakes,
to hydrothermal
or pools
activity areand
linked together
volcanic by rivers orthe
out-gassing, stream systems,
in-flow and/or containing
of solutions to hydrothermal activity
prebiotic and volcanic
feedstocks, in
combination
out-gassing, thewith
in-flowatmospheric
of solutions delivery, may prebiotic
containing provide the nutrientsinrequired
feedstocks, underwith
combination the conditions
atmospheric
essential
delivery, to orchestrate
may provide the thenutrients
first stepsrequired
towards life. Thethe
under important
conditionsrole that phosphates
essential can play in
to orchestrate thethis
first
chemistry is clear, but it is also clear that phosphate availability is highly
steps towards life. The important role that phosphates can play in this chemistry is clear, but it is dependent upon the given
environment,
also therefore, as
clear that phosphate well as phosphate
availability is highly taking an essential
dependent upon role ingiven
the orchestrating the initial
environment, steps
therefore,
of life, phosphate, by virtue of its environment dependent restrictions, could
as well as phosphate taking an essential role in orchestrating the initial steps of life, phosphate, play an essential role in
by identifying
virtue of itslikely scenariosdependent
environment and environmental
restrictions,constraints
could play forantheessential
origins role
of life. Even the most
in identifying likely
optimistic scenarios currently suggest limited amounts of phosphorus would have been soluble in
scenarios and environmental constraints for the origins of life. Even the most optimistic scenarios
the prebiotic ocean, and this may suggest that the early oceans were not the nurseries of life, rather,
currently suggest limited amounts of phosphorus would have been soluble in the prebiotic ocean,
local environments and conditions (for example, lake or river environments) could have yielded
and this may suggest that the early oceans were not the nurseries of life, rather, local environments
greater solubility of phosphorus species, through stepwise precipitation or salt leaching, and might
and conditions (for example, lake or river environments) could have yielded greater solubility of
have provided key environments for the origins of life on Earth. The unknown availability of
phosphorus species, through stepwise precipitation or salt leaching, and might have provided key
phosphates under prebiotic constraints inevitably leads to questions over their prebiotic relevance,
environments for the importance
but their biological origins of life on Earth. The
is irrefutable, unknown
moreover, the availability
simplicity ofof phosphates
fixing ammonia under
(which prebiotic
aids
constraints inevitably leads to questions over their prebiotic relevance, but
in phosphate mobilisation) through cyanide reduction, together with the valuable catalytic role that their biological importance
is irrefutable,
can be played moreover,
by simple theorganic
simplicity of fixing
amides, such asammonia
urea (8)(which aids in phosphate
and formamide, suggests thatmobilisation)
further
through cyanide reduction, together with the valuable catalytic role
investigation of the availability of phosphate needs to be considered from a systems chemistry that can be played by simple
organic amides, such as urea (8) and formamide, suggests that further
perspective [103], together with the geochemical constraints for phosphate availability. By investigation of the availability
of phosphate
proactively needsseekingtoenvironmental
be consideredconditions
from a systems
to accesschemistry
phosphate, perspective [103], together
we will undoubtedly learnwithmorethe
geochemical constraints
about the location for origin.
of life’s phosphate availability. By proactively seeking environmental conditions
to access phosphate, we will undoubtedly learn more about the location of life’s origin.
Acknowledgment: We thank the Simons Foundation (318881), the Engineering and Physical Sciences Research
Council (EP/K004980/1) and the Leverhulme Trust (RGP-2013-189) for support.
Acknowledgments: We thank the Simons Foundation (318881), the Engineering and Physical Sciences Research
Council (EP/K004980/1) and the Leverhulme Trust (RGP-2013-189) for support.
Life 2017, 7, 31 19 of 23

Conflicts of Interest: The authors declare no conflict of interest.

References
1. Khoury, G.A.; Baliban, R.C.; Floudas, C.A. Proteome-wide post-translational modification statistics:
Frequency analysis and curation of the swiss-prot database. Sci. Rep. 2011, 1, 90. [CrossRef] [PubMed]
2. Westheimer, F. Why nature chose phosphates. Science 1987, 235, 1173–1178. [CrossRef] [PubMed]
3. Kamerlin, S.C.L.; Sharma, P.K.; Prasad, R.B.; Warshel, A. Why nature really chose phosphate. Q. Rev. Biophys
2013, 46, 1–132. [CrossRef] [PubMed]
4. Goldford, J.E.; Hartman, H.; Smith, T.F.; Segrè, D. Remnants of an ancient metabolism without phosphate.
Cell 2017, 168, 1126–1134.e9. [CrossRef] [PubMed]
5. Hargreaves, W.R.; Deamer, D.W. Liposomes from ionic, single-chain amphiphiles. Biochemistry 1978, 17,
3759–3768. [CrossRef] [PubMed]
6. Szostak, J.W.; Bartel, D.P.; Luisi, P.L. Synthesizing life. Nature 2001, 409, 387–390. [CrossRef] [PubMed]
7. Luisi, P.L.; Walde, P.; Oberholzer, T. Lipid vesicles as possible intermediates in the origin of life. Curr. Opin.
Colloid Interface 1999, 4, 33–39. [CrossRef]
8. Monnard, P.-A.; Apel, C.L.; Kanavarioti, A.; Deamer, D.W. Influence of ionic inorganic solutes on
self-assembly and polymerization processes related to early forms of life: Implications for a prebiotic
aqueous medium. Astrobiology 2002, 2, 139–152. [CrossRef] [PubMed]
9. Mohammed, F.; Chen, K.; Mojica, M.; Conley, M.; Napoline, J.; Butch, C.; Pollet, P.; Krishnamurthy, R.;
Liotta, C. A plausible prebiotic origin of glyoxylate: Nonenzymatic transamination reactions of glycine with
formaldehyde. Synlett 2016, 28, 93–97. [CrossRef]
10. Bean, H.D.; Anet, F.A.L.; Gould, I.R.; Hud, N.V. Glyoxylate as a backbone linkage for a prebiotic ancestor of
RNA. Orig. Life Evol. Biol. 2006, 36, 39–63. [CrossRef] [PubMed]
11. Nelson, K.E.; Levy, M.; Miller, S.L. Peptide nucleic acids rather than RNA may have been the first genetic
molecule. Proc. Natl. Acad. Sci. USA 2000, 97, 3868–3871. [CrossRef] [PubMed]
12. Nielsen, P.E. Peptide nucleic acids and the origin of life. Chem. Biodivers. 2007, 4, 1996–2002. [CrossRef]
[PubMed]
13. Maciá, E.; Hernández, M.V.; Oró, J. Primary sources of phosphorus and phosphates in chemical evolution.
Orig. Life Evol. Biol. 1997, 27, 459–480. [CrossRef]
14. Gulick, A. Phosphorus as a factor in the origin of life. Am. Sci. 1955, 43, 479–489.
15. Gulick, A. Phosphorus and the origin of life. Ann. N. Y. Acad. Sci. 1957, 69, 309–313. [CrossRef] [PubMed]
16. Keefe, A.; Miller, S. Are polyphosphates or phosphate esters prebiotic reagents? J. Mol. Evol. 1995, 41.
[CrossRef]
17. Griffith, E.J.; Ponnamperuma, C.; Gabel, N.W. Phosphorus, a key to life on the primitive earth. Orig. Life
1977, 8, 71–85. [CrossRef] [PubMed]
18. Schwartz, A.W. Phosphorus in prebiotic chemistry. Philos. Trans. R. Soc. B 2006, 361, 1743–1749. [CrossRef]
[PubMed]
19. Holm, N.G. Glasses as sources of condensed phosphates on the early earth. Geochem. Trans. 2014, 15, 8.
[CrossRef] [PubMed]
20. Pasek, M.A.; Lauretta, D.S. Aqueous corrosion of phosphide minerals from iron meteorites: A highly reactive
source of prebiotic phosphorus on the surface of the early earth. Astrobiology 2005, 5, 515–535. [CrossRef]
[PubMed]
21. Bryant, D.E.; Kee, T.P. Direct evidence for the availability of reactive, water soluble phosphorus on the early
Earth. H-Phosphinic acid from the Nantan meteorite. Chem. Commun. 2006, 2344. [CrossRef] [PubMed]
22. Pasek, M.A.; Dworkin, J.P.; Lauretta, D.S. A radical pathway for organic phosphorylation during schreibersite
corrosion with implications for the origin of life. Geochimica et Cosmochimica Acta 2007, 71, 1721–1736.
[CrossRef]
23. Pasek, M.A.; Kee, T.P.; Bryant, D.E.; Pavlov, A.A.; Lunine, J.I. Production of potentially prebiotic condensed
phosphates by phosphorus redox chemistry. Angew. Chem. 2008, 47, 7918–7920. [CrossRef] [PubMed]
24. Keefe, A.D.; Miller, S.L. Potentially prebiotic syntheses of condensed phosphates. Orig. Life Evol. Biol. 1996,
26, 15–25. [CrossRef]
Life 2017, 7, 31 20 of 23

25. Ferris, J.P. Cyanovinyl phosphate: A prebiological phosphorylating agent? Science 1968, 161, 53–54.
[CrossRef] [PubMed]
26. Degani, C.; Halmann, M. D-glucose 1-phosphate formation by cyanogen-induced phosphorylation of
D-glucose synthesis, mechanism, and application to other reducing sugars. J. Chem. Soc. C 1971, 1459.
[CrossRef]
27. Halmann, M.; Sanchez, R.A.; Orgel, L.E. Phosphorylation of D-ribose in aqueous solution. J. Org. Chem.
1969, 34, 3702–3703. [CrossRef]
28. Lohrmann, R.; Orgel, L.E. Prebiotic synthesis: Phosphorylation in aqueous solution. Science 1968, 161, 64–66.
[CrossRef] [PubMed]
29. Powner, M.W.; Sutherland, J.D. Prebiotic chemistry: A new modus operandi. Philos. Trans. R. Soc. B 2011,
366, 2870–2877. [CrossRef] [PubMed]
30. Pasek, M.A.; Harnmeijer, J.P.; Buick, R.; Gull, M.; Atlas, Z. Evidence for reactive reduced phosphorus species
in the early Archean ocean. Proc. Natl. Acad. Sci. USA 2013, 110, 10089–10094. [CrossRef] [PubMed]
31. Gull, M.; Mojica, M.A.; Fernández, F.M.; Gaul, D.A.; Orlando, T.M.; Liotta, C.L.; Pasek, M.A. Nucleoside
phosphorylation by the mineral schreibersite. Sci. Rep. 2015, 5, 17198. [CrossRef] [PubMed]
32. Weber, A.L. Formation of pyrophosphate, tripolyphosphate, and phosphorylimidazole with the thioester, N,
S-diacetylcysteamine, as the condensing agent. J. Mol. Evol. 1981, 18, 24–29. [CrossRef] [PubMed]
33. Weber, A.L. Formation of pyrophosphate on hydroxyapatite with thioesters as condensing agents. Biosystems
1982, 15, 183–189. [CrossRef]
34. Hagan, W.J.; Parker, A.; Steuerwald, A.; Hathaway, M. Phosphate solubility and the cyanate-mediated
synthesis of pyrophosphate. Orig. Life Evol. Biol. 2007, 37, 113–122. [CrossRef] [PubMed]
35. Ferris, J.P.; Goldstein, G.; Beaulieu, D.J. Chemical evolution. IV evaluation of cyanovinyl phosphate as a
prebiotic phosphorylating agent. J. Am. Chem. Soc. 1970, 92, 6598–6603. [CrossRef]
36. Rabinowitz, J.; Chang, S.; Ponnamperuma, C. Phosphorylation on the primitive earth: Phosphorylation by
way of inorganic phosphate as a potential prebiotic process. Nature 1968, 218, 442–443. [CrossRef] [PubMed]
37. Hermes-Lima, M. Model for prebiotic pyrophosphate formation: Condensation of precipitated orthophosphate
at low temperature in the absence of condensing or phosphorylating agents. J. Mol. Evol. 1990, 31, 353–358.
[CrossRef]
38. Yamagata, Y.; Watanabe, H.; Saitoh, M.; Namba, T. Volcanic production of polyphosphates and its relevance
to prebiotic evolution. Nature 1991, 352, 516–519. [CrossRef] [PubMed]
39. Osterberg, R.; Orgel, L.E. Polyphosphate and trimetaphosphate formation under potentially prebiotic
conditions. J. Mol. Evol. 1972, 1, 241–248. [CrossRef] [PubMed]
40. Handschuh, G.J.; Orgel, L.E. Struvite and prebiotic phosphorylation. Science 1973, 179, 483–484. [CrossRef]
[PubMed]
41. Gull, M.; Pasek, M. Is struvite a prebiotic mineral? Life 2013, 3, 321–330. [CrossRef] [PubMed]
42. Handschuh, G.J.; Lohrmann, R.; Orgel, L.E. The effect of Mg2+ and Ca2+ on urea-catalyzed phosphorylation
reactions. J. Mol. Evol. 1973, 2, 251–262. [CrossRef] [PubMed]
43. Schwartz, A.; Ponnamperuma, C. Phosphorylation on the primitive earth: Phosphorylation of adenosine
with linear polyphosphate salts in aqueous solution. Nature 1968, 218, 443. [CrossRef] [PubMed]
44. Schwartz, A.W. Specific phosphorylation of the 20 - and 30 - positions in ribonucleosides. J. Chem. Soc. D 1969,
1393a. [CrossRef]
45. Ozawa, K.; Nemoto, A.; Imai, E.; Honda, H.; Hatori, K.; Matsuno, K. Phosphorylation of nucleotide molecules
in hydrothermal environments. Orig. Life Evol. Biol. 2004, 34, 465–471. [CrossRef]
46. Yamagata, Y.; Inoue, H.; Inomata, K. Specific effect of magnesium ion on 20 , 30 -cyclic amp synthesis from
adenosine and trimeta phosphate in aqueous solution. Orig. Life Evol. Biol. 1995, 25, 47–52. [CrossRef]
47. Saffhill, R. Selective phosphorylation of the cis-2’, 3’-diol of unprotected ribonucleosides with
trimetaphosphate in aqueous solution. J. Org. Chem. 1970, 35, 2881–2883. [CrossRef] [PubMed]
48. Rabinowitz, J. Recherches sur la formation et la transformation des esters LXXXIII [1]. Réactions de
condensation et/ou de phosphorylation, en solution aqueuse, de divers composés organiques à fonctions
-OH, COOH, NH2, ou autres, à l’aide de polyphosphates linéaire. Helv. Chim. Acta 1969, 52, 2663–2671.
[CrossRef]
49. Rabinowitz, J. Peptide and amide bond formation in aqueous solutions of cyclic or linear polyphosphates as
a possible prebiotic process. Helv. Chim. Acta 1970, 53, 1350–1355. [CrossRef] [PubMed]
Life 2017, 7, 31 21 of 23

50. Rabinowitz, J.; Flores, J.; Krebsbach, R.; Rogers, G. Peptide formation in the presence of linear or cyclic
polyphosphates. Nature 1969, 224, 795–796. [CrossRef] [PubMed]
51. Tsuhako, M.; Fujimoto, M.; Ohashi, S.; Nariai, H.; Motooka, I. Phosphorylation of nucleosides with sodium
cyclo -triphosphate. Bull. Chem. Soc. Jpn. 1984, 57, 3274–3280. [CrossRef]
52. Chung, N.M.; Lohrmann, R.; Orgel, L.E.; Rabinowitz, J. The mechanism of the trimetaphosphate-induced
peptide synthesis. Tetrahedron 1971, 27, 1205–1210. [CrossRef]
53. Inoue, H.; Baba, Y.; Furukawa, T.; Maeda, Y.; Tsuhako, M. Formation of dipeptide in the reaction of amino
acids with cyclo-triphosphate. Chem. Pharm. Bull. 1993, 41, 1895–1899. [CrossRef]
54. Tsuhako, M.; Nakajima, A.; Miyajima, T.; Ohashi, S.; Nariai, H.; Motooka, I. The reaction of cyclo-triphosphate
with L-α- or β-alanine. Bull. Chem. Soc. Jpn. 1985, 58, 3092–3098. [CrossRef]
55. Gao, X.; Liu, Y.; Xu, P.X.; Cai, Y.M.; Zhao, Y.F. α-Amino acid behaves differently from β- and γ-amino acids
as treated by trimetaphosphate. Amino Acids 2008, 34, 47–53. [CrossRef] [PubMed]
56. Quimby, O.T.; Flautt, T.J. Ammonolyse des trimetaphosphats. Z. Anorg. Allg. Chem. 1958, 296, 220–228.
[CrossRef]
57. Feldmann, W.; Thilo, E. Zur chemie der kondensierten phosphate und arsenate. XXXVIII. amidotriphosphat.
Z. Anorg. Allg. Chem. 1964, 328, 113–126. [CrossRef]
58. Jiang, Y.; Tan, B.; Chen, Z.Z.; Liu, T.; Zhong, R.G.; Li, Y.M.; Stewart, D.J.; Zhao, Y.F.; Jiang, H.L. Phosphoryl
group differentiating α-amino acids from β- and γ-amino acids in prebiotic peptide formation. Int. J.
Quantum Chem. 2003, 94, 232–241. [CrossRef]
59. Yamanaka, J.; Inomata, K.; Yamagata, Y. Condensation of oligoglycines with trimeta- and tetrametaphosphate
in aqueous solutions. Orig. Life Evol. Biol. 1988, 18, 165–178. [CrossRef]
60. Kolb, V.; Orgel, L.E. Phosphorylation of glyceric acid in aqueous solution using trimetaphosphate. Orig. Life
Evol. Biol. 1996, 26, 7–13. [CrossRef]
61. Kolb, V.; Zhang, S.; Xu, Y.; Arrhenius, G. Mineral induced phosphorylation of glycolate ion—A metaphor in
chemical evolution. Orig. Life Evol. Biol. 1997, 27, 485–503. [CrossRef]
62. Feldmann, W. Zur chemie der kondensierten phosphate und arsenate, LIII. das trimetaphosphat als
triphosphorylierungsmittel für alkohole und kohlenhydrate in wäßriger Lösung. seine sonderstellung
unter den kondensierten phosphaten. Chem. Ber. 1967, 100, 3850–3860. [CrossRef]
63. Müller, D.; Pitsch, S.; Kittaka, A.; Wagner, E.; Wintner, C.E.; Eschenmoser, A.; Ohlofjgewidmet, G. Chemie von
a-aminonitrilen. aldomerisierung von glycolaldehyd-phosphat zu racemischen hexose-2,4,6-triphosphaten
und (in gegenwart von formaldehyd) racemischen pentose-2,4-diphosphaten: Rac-allose-2,4,6-triphosphat
und rac-ribose-2,4-diphosphat sind die reaktionshauptprodukte. Helv. Chim. Acta 1990, 73, 1410–1468.
[CrossRef]
64. Krishnamurthy, R.; Arrhenius, G.; Eschenmoser, A. Formation of glycolaldehyde phosphate from
glycolaldehyde in aqueous solution. Orig. Life Evol. Biol. 1999, 29, 333–354. [CrossRef]
65. Pitsch, S.; Eschenmoser, A.; Gedulin, B.; Hui, S.; Arrhenius, G. Mineral induced formation of sugar
phosphates. Orig. Life Evol. Biol. 1995, 25, 297–334. [CrossRef]
66. Mullen, L.B.; Sutherland, J.D. Formation of potentially prebiotic amphiphiles by reaction of
β-hydroxy-n-alkylamines with cyclotriphosphate. Angew. Chem. 2007, 46, 4166–4168. [CrossRef] [PubMed]
67. Krishnamurthy, R.; Guntha, S.; Eschenmoser, A. Regioselective α-phosphorylation of aldoses in aqueous
solution. Angew. Chem. 2000, 39, 2281–2285. [CrossRef]
68. Coggins, A.J.; Powner, M.W. Prebiotic synthesis of phosphoenol pyruvate by α-phosphorylation-controlled
triose glycolysis. Nat. Chem. 2017, 9, 310–317. [CrossRef] [PubMed]
69. Pitsch, S.; Pombo-Villar, E.; Eschenmoser, A. chemie von α-aminonitrilen. 13. mitteilung. über die bildung
von 2-oxoethyl-phosphaten (“glycoladehyd-phosphaten”) aus rac-oxirancarbonitril und anorganischem
phosphat und über (formale) konstitutionelle zusammenhänge zwischen 2-oxoethyl-phosphaten und oligo
(hexo- und pentopyranosyl)nucleotid-rückgraten. Helv. Chim. Acta 1994, 77, 2251–2285. [CrossRef]
70. Wagner, E.; Xiang, Y.B.; Baumann, K.; Gück, J.; Eschenmoser, a. chemie von α-aminonitrilen. aziridin-2-carbonitril,
ein vorläufer von rca-o3-phosphoserinnitril und glycolaldehyd-phosphat. Helv. Chim. Acta 1990, 73,
1391–1409. [CrossRef]
71. Sanchez, R.A.; Orgel, L.E. Studies in prebiotic synthesis. J. Mol. Biol. 1970, 47, 531–543. [CrossRef]
72. Anastasi, C.; Crowe, M.A.; Sutherland, J.D. Two-step potentially prebiotic synthesis of α-d-cytidine-5’-
phosphate from d-glyceraldehyde-3-phosphate. J. Am. Chem. Soc. 2007, 129, 24–25. [CrossRef] [PubMed]
Life 2017, 7, 31 22 of 23

73. Richard, J.P. Acid-base catalysis of the elimination and isomerization reactions of triose phosphates. J. Am.
Chem. Soc. 1984, 106, 4926–4936. [CrossRef]
74. Gauss, D.; Schoenenberger, B.; Wohlgemuth, R. Chemical and enzymatic methodologies for the synthesis of
enantiomerically pure glyceraldehyde 3-phosphates. Carbohydr. Res. 2014, 389, 18–24. [CrossRef] [PubMed]
75. Fernández-García, C.; Grefenstette, N.M.; Powner, M.W. Prebiotic synthesis of aminooxazoline-50 -phosphates
in water by oxidative phosphorylation. Chem. Commun. 2017, 53, 4919–4921. [CrossRef] [PubMed]
76. Hargreaves, W.R.; Mulvihill, S.J.; Deamer, D.W. Synthesis of phospholipids and membranes in prebiotic
conditions. Nature 1977, 266, 78–80. [CrossRef] [PubMed]
77. Maheen, G.; Tian, G.; Wang, Y.; He, C.; Shi, Z.; Yuan, H.; Feng, S. Resolving the enigma of prebiotic C-O-P
bond formation: Prebiotic hydrothermal synthesis of important biological phosphate esters. Heteroat. Chem.
2010, 21, 161–167. [CrossRef]
78. Epps, D.E.; Sherwood, E.; Eichberg, J.; Oró, J. Cyanamide mediated syntheses under plausible primitive
earth conditions. J. Mol. Evol. 1978, 11, 279–292. [CrossRef] [PubMed]
79. Ponnamperuma, C.; Mack, R. Nucleotide synthesis under possible primitive earth conditions. Science 1965,
148, 1221–1223. [CrossRef] [PubMed]
80. Škoda, J.; Morávek, J. Formation of uridylyl (30 → 50 ) uridine, uridylyl (20 → 50 ) uridine, 6-azauridylyl (30 →
50 )6-azauridine and 6-azauridylyl (20 → 50 )6-azauridine by thermic phosphorylation of the corresponding
nucleosides with inorganic phosphate. Tetrahedron Lett. 1966, 7, 4167–4172. [CrossRef]
81. Beck, A.; Lohrmann, R.; Orgel, L.E. Phosphorylation with inorganic phosphates at moderate temperatures.
Science 1967, 157, 952. [CrossRef] [PubMed]
82. Österberg, R.; Orgel, L.E.; Lohrmann, R. Further studies of urea-catalyzed phosphorylation reactions.
J. Mol. Evol. 1973, 2, 231–234. [CrossRef] [PubMed]
83. Lohrmann, R.; Orgel, L.E. Urea-inorganic phosphate mixtures as prebiotic phosphorylating agents. Science
1971, 171, 490–494. [CrossRef] [PubMed]
84. Bishop, M.J.; Lohrmann, R.; Orgel, L.E. Prebiotic phosphorylation of thymidine at 65 ◦ C in simulated desert
conditions. Nature 1972, 237, 162–164. [CrossRef] [PubMed]
85. Reimann, R.; Zubay, G. Nucleoside phosphorylation: A feasible step in the prebiotic pathway to RNA.
Orig. Life Evol. Biol. 1999, 29, 229–247. [CrossRef]
86. Stairs, S.; Nikmal, A.; Bučar, D.K.; Zheng, S.L.; Szostak, J.W.; Powner, M.W. Divergent prebiotic synthesis of
pyrimidine and 8-oxo-purine ribonucleotides. Nat. Commun. 2017, 8, 15270. [CrossRef] [PubMed]
87. Powner, M.W.; Gerland, B.; Sutherland, J.D. Synthesis of activated pyrimidine ribonucleotides in prebiotically
plausible conditions. Nature 2009, 459, 239–242. [CrossRef] [PubMed]
88. Choudhary, A.; Kamer, K.J.; Powner, M.W.; Sutherland, J.D.; Raines, R.T. A stereoelectronic effect in prebiotic
nucleotide synthesis. ACS Chem. Biol. 2010, 5, 655–657. [CrossRef] [PubMed]
89. Patel, B.H.; Percivalle, C.; Ritson, D.J.; Duffy, C.D.; Sutherland, J.D. Common origins of RNA, protein and
lipid precursors in a cyanosulfidic protometabolism. Nat. Chem. 2015, 7, 301–307. [CrossRef] [PubMed]
90. Philipp, M.; Seliger, H. Spontaneous phosphorylation of nucleosides in formamide-ammonium phosphate
mixtures. Naturwissenschaften 1977, 64, 273. [CrossRef] [PubMed]
91. Schoffstall, A.M.; Barto, R.J.; Ramos, D.L. Nucleoside and deoxynucleoside phosphorylation in formamide
solutions. Orig. Life 1982, 12, 143–151. [CrossRef] [PubMed]
92. Schoffstall, A.M.; Laing, E.M. Phosphorylation mechanisms in chemical evolution. Orig. Life Evol. Biol. 1985,
15, 141–150. [CrossRef]
93. Cockerill, A.F.; Deacon, A.; Harrison, R.G.; Osborne, D.J.; Prime, D.M.; Ross, W.J.; Todd, A.; Verge, J.P.
An improved synthesis of 2-amino-1,3-oxazoles under basic catalysis. Synthesis 1976, 1976, 591–593.
[CrossRef]
94. Anastasi, C.; Crowe, M.A.; Powner, M.W.; Sutherland, J.D. Direct assembly of nucleoside precursors from
two- and three-carbon units. Angew. Chem. 2006, 45, 6176–6179. [CrossRef] [PubMed]
95. Powner, M.W.; Sutherland, J.D. Phosphate-mediated interconversion of ribo- and arabino-configured prebiotic
nucleotide intermediates. Angew. Chem. 2010, 49, 4641–4643. [CrossRef] [PubMed]
96. Ingar, A.A.; Luke, R.W. A.; Hayter, B.R.; Sutherland, J.D. Synthesis of cytidine ribonucleotides by stepwise
assembly of the heterocycle on a sugar phosphate. Chem. Biol. Chem. 2003, 4, 504–507. [CrossRef] [PubMed]
97. Powner, M.W.; Zheng, S.L.; Szostak, J.W. Multicomponent assembly of proposed DNA precursors in water.
J. Am. Chem. Soc. 2012, 134, 13889–13895. [CrossRef] [PubMed]
Life 2017, 7, 31 23 of 23

98. Islam, S.; Bučar, D.K.; Powner, M.W. Prebiotic selection and assembly of proteinogenic amino acids and
natural nucleotides from complex mixtures. Nat. Chem. 2017, 9, 584–589. [CrossRef]
99. Coggins, A.J.; Tocher, D.A.; Powner, M.W. One-step protecting-group-free synthesis of azepinomycin in
water. Org. Biomol. Chem. 2015, 13, 3378–3381. [CrossRef] [PubMed]
100. Li, L.; Prywes, N.; Tam, C.P.; O’Flaherty, D.K.; Lelyveld, V.S.; Izgu, E.C.; Pal, A.; Szostak, J.W. Enhanced
nonenzymatic RNA copying with 2-aminoimidazole activated nucleotides. J. Am. Chem. Soc. 2017, 139,
1810–1813. [CrossRef] [PubMed]
101. Fahrenbach, A.C.; Giurgiu, C.; Tam, C.P.; Li, L.; Hongo, Y.; Aono, M.; Szostak, J.W. Common and potentially
prebiotic origin for precursors of nucleotide synthesis and activation. J. Am. Chem. Soc. 2017. [CrossRef]
[PubMed]
102. Patel, A.; Malinovska, L.; Saha, S.; Wang, J.; Alberti, S.; Krishnan, Y.; Hyman, A.A. ATP as a biological
hydrotrope. Science 2017, 356, 753–756. [CrossRef] [PubMed]
103. Islam, S.; Powner, M.W. Prebiotic systems chemistry: Complexity overcoming clutter. Chem 2017, 2, 470–501.
[CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).