The Mechanism of Cleavage of RNA Phosphodiesters by a Gold Nanoparticle Nanozyme

Article abstract of DOI:10.1002/chem.202100299

The cleavage of uridine 3’-phosphodiesters bearing alcohols with pK_a ranging from 7.14 to 14.5 catalyzed by AuNPs functionalized with 1,4,7-triazacyclononane-Zn(II) complexes has been studied to unravel the source of catalysis by these nanosystems (nanozymes). The results have been compared with those obtained with two Zn(II) dinuclear catalysts for which the mechanism is fairly understood. Binding to the Zn(II) ions by the substrate and the uracil of uridine was observed. The latter leads to inhibition of the process and formation of less productive binding complexes than in the absence of the nucleobase. The nanozyme operates with these substrates mostly via a nucleophilic mechanism with little stabilization of the pentacoordinated phosphorane and moderate assistance in leaving group departure. This is attributed to a decrease of binding strength of the substrate to the catalytic site in reaching the transition state due to an unfavorable binding mode with the uracil. The nanozyme favors substrates with better leaving groups than the less acidic ones.

Full text of DOI:10.1002/chem.202100299

Full Paper
doi.org/10.1002/chem.202100299
Chemistry—A European Journal
The Mechanism of Cleavage of RNA Phosphodiesters by a
Gold Nanoparticle Nanozyme
Joanna Czescik,[a, c] Fabrizio Mancin,[a] Roger Strömberg,*[b] and Paolo Scrimin*[a]
Abstract: The cleavage of uridine 3’-phosphodiesters bearing
alcohols with pKa ranging from 7.14 to 14.5 catalyzed by AuNPs
functionalized with 1,4,7-triazacyclononane-Zn(II) complexes has
been studied to unravel the source of catalysis by these
nanosystems (nanozymes). The results have been compared
with those obtained with two Zn(II) dinuclear catalysts for which
the mechanism is fairly understood. Binding to the Zn(II) ions by
the substrate and the uracil of uridine was observed. The latter
leads to inhibition of the process and formation of less
productive binding complexes than in the absence of the
nucleobase. The nanozyme operates with these substrates
mostly via a nucleophilic mechanism with little stabilization of
the pentacoordinated phosphorane and moderate assistance in
leaving group departure. This is attributed to a decrease of
binding strength of the substrate to the catalytic site in reaching
the transition state due to an unfavorable binding mode with
the uracil. The nanozyme favors substrates with better leaving
groups than the less acidic ones.
Introduction
precise role of each contribution is a challenging task due to the
interconnectedness of each process. By summing-up all of them,
rate accelerations approaching 15–16 orders of magnitude have
been estimated, not much different from those observed for
nucleases.[5,6]
The non-enzymatic hydrolytic cleavage of phosphate diesters is a
continuously challenging endeavor for scientists because of their
key role in connecting the constituent nucleosides of DNA and
RNA.[1] Efficient catalysts could find an application in the fields of
biotechnology, chemotherapy, and medicine. In the absence of
catalysts, the phosphodiester linkage is exceptionally stable under
physiological conditions.[2] Nevertheless, natural enzymes (nucle-
ases) perform this job with high efficiency.[3] In the case of RNA,
metal ions are often present in the catalytic site of these enzymes
and play quite important roles in the catalytic process.[4] For
instance, they prevent the electrostatic repulsion between
negatively charged species, they are involved in the Lewis acid
activation of the electrophile, in the increase of the concentration
of deprotonated nucleophiles at neutral pH, and in facilitating the
departure of the leaving group via pKa reduction.[5] Elucidating the
Numerous studies have shown that bimetallic catalysts are far
better than their mononuclear counterparts.[7] Two metal ions may
perform complementary tasks and optimize catalyst-substrate
interactions. Multimetallic catalysts are more complex multivalent
systems,[8] typically showing enhanced activity because of the
introduction of a statistical amplification of the interactions with a
substrate.[9] Metal ion-functionalized gold nanoparticles (AuNPs)
belong to this category. Indeed, they proved to be rather efficient
catalysts for the cleavage of phosphodiesters, including RNA
model substrates, and DNA as well.[10] They are self-assembled
supramolecular systems in which a cluster of gold atoms is coated
with a passivating monolayer, which, for this application, is
functionalized with metal ion complexes.[11] For their mode of
action, and catalytic efficiency (up to 5–6 orders of magnitude rate
acceleration with model substrates) they have been dubbed
nanozymes[12] and are among the most efficient catalysts reported
so far.
[a] Dr. J. Czescik, Prof. Dr. F. Mancin, Prof. Dr. P. Scrimin
Department of Chemical Sciences
University of Padova
Via Marzolo, 1 – 35131 Padova (Italy)
E-mail: paolo.scrimin@unipd.it
Commonly utilized, simple RNA model substrates are usually
characterized by the presence of efficient leaving groups and are
devoid of the nucleobases. The most used one is 2-hydroxypropyl
p-nitrophenyl phosphate, HPNP (Figure 1).[13] Good leaving groups
require very little (if any) protonation to depart from the substrate
contrary to phosphate diesters of nucleic acids bearing a poorly
acidic 5’-nucleoside.
[b] Prof. Dr. R. Strömberg
Karolinska Institutet
NEO, 14157 Huddinge (Sweden)
E-mail: roger.stromberg@ki.se
[c] Dr. J. Czescik
Current address:
School of Life and Health Sciences
Aston University
In order to assess the efficiency of an artificial catalyst more
challenging substrates must be studied, including those with a
poor leaving group.[14]
A universal, accepted mechanism does not appear to exist for
the metal ion-promoted cleavage of RNA model substrates.[15] The
mechanistic studies are also complicated by the fact that similar
kinetic evidence can concurrently support different mechanisms.
B4 7ET Birmingham (UK)
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202100299
© 2021 The Authors. Chemistry - A European Journal published by Wiley-
VCH GmbH. This is an open access article under the terms of the Creative
Commons Attribution Non-Commercial NoDerivs License, which permits use
and distribution in any medium, provided the original work is properly cited,
the use is non-commercial and no modifications or adaptations are made.
Chem. Eur. J. 2021, 27, 1–7
1
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100299
Chemistry—A European Journal
Cooperativity between two Zn(II) complexes has been shown to
be favored by the presence of the amide group in the tether
connecting TACN to the gold cluster.[13,20] In the passivating
monolayer N-H…O=C H-bonds are formed leading to a more
tightly packed structure.[20,21] Furthermore, the hydrocarbon chain
ensures a hydrophobic environment resembling that of enzymatic
active sites, which stabilizes the electrostatic interactions as well as
desolvates nucleophiles. In the kinetic studies, the concentration
of nanoparticles reported refers to the concentration of the TACN
ligand present on their monolayer. This allowed us to evaluate the
role of the metal complexes independently of their number on
each nanoparticle.
To assess the difference in behavior between HPNP and
uridine 3’-phosphodiesters, we first ran kinetics using the very
reactive uridine 3’-p-nitrophenyl phosphate (UPNP). Under Mi-
chaelis-Menten conditions (i.e. excess substrate over catalyst), to
determine the relevant kinetic parameters, we obtained the graph
reported in Figure 2A. Instead of showing the saturation profile
typical of an enzyme-like catalyst, the rate reached a maximum
Figure 1. Catalysts and substrates discussed in this work.
Indeed, metal ions can enhance both the nucleophilic attack and
leaving group departure following a general base or acid catalysis
or even a bifunctional general base-general acid catalysis. More-
over, the mechanism also depends on the acidity of the metal-
coordinated water molecule, the strength of the metal ion-
substrate complex and the nature of the leaving group.[16]
Accordingly, subtle changes within these parameters can signifi-
cantly alter the resulting mechanism and, ultimately, the reaction
rate. In addition, nucleobase-functionalized phosphates (as those
present in RNA and DNA) may influence the interaction with the
catalyst taking advantage of the ability of some of them to interact
with the metal ion possibly altering the conformation of the
complex.[17]
Our previous studies on HPNP hydrolysis catalyzed by AuNPs
functionalized with 1,4,7-triazacyclononane, TACN, Zn(II) com-
plexes revealed that the catalytic contribution could be attributed
to nucleophilic activation as well as to Lewis acid catalysis.[13,18,19]
Despite being one of the best catalysts reported for phosphate
diester cleavage, no study regarding the role of the leaving group
departure in AuNPs catalysis has been reported so far. Herein we
report our studies on the hydrolysis of uridine 3’-phosphodiesters
bearing alcohols with pKa ranging from 7.14 to 14.5 using as the
catalyst, the best Zn(II)-based nanozyme reported to date, for the
cleavage of the RNA model substrate, HPNP.[13] We have also
investigated how the uracil moiety present in the substrate affects
its interaction with the catalyst and, ultimately, its catalytic
efficiency. Our results were additionally compared with those
reported for other dinuclear Zn(II)-based catalysts to unravel
similarities and differences between them and our nanoparticles.
The aim was to obtain a more comprehensive understanding of
the mechanism of the reaction and elucidate the source of
catalysis with these nanozymes.
Results and Discussion
The transesterification of UPNP
Figure 2. A. Initial rate vs UPNP concentration profile for the cleavage
process catalyzed by AuNP1. The dotted line represents the fitting of the
first four points with the Michaelis-Menten equation. Conditions: [AuNP1]=
The gold nanoparticles we have used for our study (AuNP1,
Figure 1) feature in the coating monolayer, a terminal TACN, as
the ligand for Zn(II) complexation in situ. The mode of action of
this catalyst requires the cooperation between two metal ions.[18,19]
[Zn(II)]=5.0×10À 6 M, [HEPES]= 0.01 M, pH 7.5, 25 C. B. Observed rate
°
constant for the cleavage of UPNP at increasing AuNP1 concentration. The
dotted line is meant exclusively to guide the eye. Conditions: [AuNP1]=[Zn-
(II)], [HEPES]=0.01 M, [UPNP]=3.0×10À 5 M, pH 7.5, 25 C.
°
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
2
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100299
Chemistry—A European Journal
followed by a relatively sharp decrease. Thus, the curve obtained
for UPNP indicates that, in the presence of AuNP1 (at [AuNP1]=
5.0×10À 6 M), when [UPNP] >3.7×10À 5 M, the substrate starts
inhibiting the reaction. The same phenomenon can be observed
by running kinetics at constant UPNP concentration and increas-
ing [AuNP1]. In this case (Figure 2B) the graph shows and upward
curvature indicative of enhanced performance as the [catalyst]/
[substrate] ratio increases. From the Michaelis-Menten analysis of
the first portion of the curve (up to [UPNP]=37 μM) for UPNP
cleavage catalyzed by AuNP1 (Figure 2A) we obtained the relevant
kinetic parameters reported in Table 1. Although this analysis
should be taken with some caution due to the competing
inhibition process, some interesting observations can be made.
First, kcat is higher for HPNP than for UPNP (ca 13-fold).
Second, the krel (kcat/kuncat) values indicate a 34-fold acceleration
of the cleavage of UPNP vs almost one million-fold for HPNP
compared to the spontaneous, uncatalyzed process for the two
substrates. Nevertheless, k2 (kcat/KM) is higher for UPNP than for
HPNP. The better performance of UPNP, in terms of second order
rate constant, derives from a far lower KM for this substrate that
amounts to an almost 20-fold higher binding constant to the
monolayer compared to that of HPNP. The above data suggest
that UPNP binds to the TACNÀ Zn(II) complexes on the nano-
particle not only through the coordination of the phosphate but
also through the uridine moiety. Binding of uridine to Zn(II)
complexes is well known and occurs through the deprotonated
imide nitrogen.[17] Lönnberg has shown that polynuclear catalytic
complexes interact with the uridines of UpU and that at least a
third metal ion is necessary to cleave the phosphate in a 1:1
complex.[22] This double interaction occurring on the nanoparticle
surface increases the overall binding constant. However, the
binding of the nucleobase to a nearby Zn(II) complex makes it
unavailable for cooperating in the cleavage of the substrate. We
have shown that phosphate diesters (but not triesters)[23] cleavage
substrate, inhibition by dimethylphosphate leads to a Ki =7.6×
10À 4 M. The effective molarity of the intramolecular binding
process, due to the contemporary presence of the phosphate and
uridine, EM= Ki,DMP ×Ki,uridine/KM, is ca. 33 mM. This is less than what
predicted by Mandolini for positive cooperativity (>60 mM).[25]
This implies a non-optimal binding mode of UPNP to AuNP1 not
allowing to take full advantage of both binding units.
Richard and Morrows,[26,27] demonstrated enhanced catalytic
efficiency of dinuclear catalyst 1 (Figure 1) for the cleavage of
HPNP over UPNP (with respect to the uncatalyzed process) and
concluded that facile access to the cationic catalytic site is sterically
blocked for the bulkier UPNP substrate, an unlikely scenario for
the nanoparticles because of the flexibility of the monolayer. In
order to better understand the source of catalysis with AuNP1, we
investigated the temperature dependence of the reaction rate for
the cleavage of UPNP and compared the thermodynamic results
with those available for HPNP (Table 2).[18] The analysis of Table 2
reveals that, while the entropic contribution to ΔG� is lower for
UPNP, the enthalpic contribution is higher. The lower -ΔS� is likely
the result of the absence of conformational freedom of the 2’-OH
of UPNP locked in the tetrahydrofuran ring. On the contrary, the
hydroxypropyl unit of HPNP can freely rotate thus populating also
non-productive conformations. The movement towards the tran-
sition state for this substrate requires freezing of such a rotation.
The higher ΔH� could result from a partial detachment of UPNP
from the bimetallic catalytic site connected to a non-appropriate
geometry of binding associated with the presence of the uridine
moiety as suggested also by the low EM determined above. Thus,
for AuNP1 catalysis, although the interaction of UPNP with the
nanoparticle is stronger than that with HPNP, the formed complex
is not the productive one leading to an efficient transesterification
reaction. In order to properly orient the phosphate for the attack
by the 2’-OH, part of that binding energy gain must be lost in the
transition state. The stronger binding of UPNP to AuNP1 also
affects the pH vs rate profile of the reaction. Figure 3 shows that
the rate of the catalyzed reaction goes up linearly with the
increase of pH with slope=1 and starts flattening at pH>8. We
have analyzed the curve considering one acidity constant
(Equation 1):
by TACNÀ Zn(II) complexes requires
a
dinuclear catalytic
site.[10,12,18,19] As the concentration of UPNP increases, dinuclear
catalytic sites become less and less available and the catalytic
performance of the catalyst decreases (downward curvature of the
graph of Figure 2A). The ability of uridine to bind to the
monolayer could be confirmed in inhibition experiments using
HPNP as the substrate. The Ki obtained (1.3×10À 3 M, see
Supporting Information) indicates that uridine binds to the Zn(II)
complexes on the nanoparticle although with a slightly weaker
binding constant than a phosphate diester.[24] With the same
kobs ¼ k2½AuNP1�ðKa=Ka þ ½Hþ�Þ
(1)
The best fitting gave a pKa of 8.35 for the acidic species
involved in the catalytic process, either a water molecule (base
catalysis) or 2’-OH (nucleophilic catalysis) coordinated to Zn(II). In
the case of HPNP the analogous pH vs rate profile is bell-shaped
with two pKa involved (7.8 and 9.2).[13] The second pKa is typically
associated with the deprotonation of a second Zn(II)-bound water
Table 1. Reactivity parameters for the AuNP1 catalyzed cleavage of UPNP
and HPNP.[a]
Substrate
UPNP[b]
HPNP[c]
kcat, sÀ 1
0.0146
2.98×10À 5
490
0.193
KM, M
5.80×10À 4
333
k2 (kcat/KM), MÀ 1 sÀ 1
Table 2. Thermodynamic parameters for AuNP1 catalyzed cleavage of
HPNP and UPNP.
krel =kcat/kuncat
34[d]
9.7×105[e]
Substrate[a]
ΔH�, KJmolÀ 1
ΔS�, JmolÀ 1
ΔG�, KJmolÀ 1[b]
[a] Conditions for UPNP cleavage: [AuNP1]=[Zn(II)]=5.0 x 10À 6 M, [HEPES]
°
=
0.01 M, pH 7.5, 25 C; for HPNP cleavage: [AuNP1]=[Zn(II)]=2.5×
HPNP
UPNP
64.5
76.2
À 53.3
À 28.4
81.1
85.1
10À 5 M, [EPPS, buffer]=1.0×10À 2 M, pH=7.5, 25 C. [b] The analysis was
°
performed on the first four points of the graph of Figure 2A. [c] Data from
ref. 13. [d] kuncat =4.3×10À 4 sÀ 1. [e] kuncat =2.0×10À 7 sÀ 1
.
[a] Data for HPNP from ref. 18, for UPNP this work. [b] T=313 K.
°
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
3
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100299
Chemistry—A European Journal
formation of the cyclic phosphate (2’,3’-cUMP) which eventually
formed two uridine monophosphates (2’-UMP and 3’-UMP), see
(a) in Scheme 1, with a clear preference for 2’OÀ P over 3’OÀ P
bond cleavage. Analogously to what was reported for 1,[28] no
evidence of isomerization of the uridine 3’-phosphate diesters into
the isomeric 2’-phosphates was obtained. This indicates that over
the course of the transesterification process there is not enough
stabilization of the transient cyclic phosphorane (Scheme 1 (b)) as,
on the contrary, reported by Williams et al. for the dinuclear
catalyst 2.[29] This suggests that for AuNP1 the conversion of the
starting uridine 3’-phosphate diester into 2’,3’-cUMP involves a
single step in which nucleophile addition and leaving group
departure are concerted (although not synchronous). The Brønsted
°
plot, obtained at pH=7.5 and 40 C, reported in Figure 4 is linear
in the entire interval of pKa of the leaving groups explored (7.14–
12.89) and gives a βlg =À 0.86. The same plot for catalysts 1 (pKa
interval 8.35–14.31), obtained by running the kinetics under the
same conditions, is equally linear and gives βlg =À 0.59 (Figure 4).
This value is slightly lower than those reported for this catalyst
under different conditions (À 0.72 by Richard et al.28] and À 0.63 or
À 0.74 by Mikkola et al.[15] for aromatic and aliphatic leaving
groups, respectively). The absence of any break in the linear plot
for catalysts 1 and AuNP1 indicates there is no change of
mechanism for both of them. Williams et al.[29] reported, for a
limited set of pKa for catalyst 2, a possible deviation from linearity
at pKa ca. 12.4, not significantly different from the value at which
the break point for the OH-catalyzed process was reported.[30] At
Figure 3. Dependence of the rate constant (k2 for HPNP and kobs for UPNP)
from pH for the cleavage of HPNP and UPNP by AuNP1. The dotted lines
refer to the interpolation of the points with two protonation constants for
HPNP, red symbols, and a single protonation constant for UPNP, blue
symbols. Data for HPNP from ref 13. Conditions for UPNP: [AuNP1]-
=5×10À 6 M, [UPNP]=3.0×10À 5 M, [buffers]=1.0×10À 2 M, T=25 C. Note
°
that since the rate constants reported for the two substrates are different
the two curves cannot be used to compare relative reactivities but only the
profile of the pH dependence.
molecule. The resulting OH- is strongly bound to the metal ion
thus inhibiting the binding of the substrate. The more tightly
bound UPNP competes more favorably with the OH- for the
coordination to Zn(II) and remains attached to the catalytic site
also at high pH. Notably, catalyst 1 shows a single, catalytically
relevant pKa (7.8) in the cleavage of HPNP. Likely this is due to a
higher pKa for the water molecule bound to the second Zn(II) ion
that falls beyond the pH interval studied.
The role of the pKa of the leaving group
Scheme 1. (a) Products and intermediate formed in the cleavage of
phosphates 3 by AuNP1. (b) Cleavage of uridine 3’-phosphates with the
formation of a phosphorane intermediate.
In addition to UPNP, we have studied five more uridine 3’-alkyl-
and aryl-phosphodiesters (3a–e, Figure 1) to assess the role of the
basicity of the leaving group in the AuNP1-catalyzed trans-
°
esterification process. Reactions were performed at 40 C and
Table 3. Bronsted parameters for the cleavage of uridine 3’-phosphate
diesters in the presence of different catalysts.
pH 7.5 in the presence of 5.0 μM AuNP1.The kinetics with the least
reactive substrates (3b–d) could be followed by HPLC. In the case
of 3a the reaction was so slow that AuNP1 decomposition
became important and no reliable kinetic data could be obtained
for this catalyst. The use of HPLC allowed us to follow the course
of the reaction and monitor the final and intermediate products
formed. Reaction rates for the far more reactive UPNP and 3e
substrates were determined from the extrapolation of the
Arrhenius plots obtained from kinetics run at lower temperature
and followed spectrophotometrically (see the Supporting Informa-
tion). Analysis of the HPLC traces revealed the disappearance of
the original uridine 3’-phosphate diesters, the intermediate
Catalyst βlg (aromatic leaving
βlg (aliphatic leaving
groups)
Reference
groups)
OH-
Zn(II)
AuNP1
1
À 1.34[a]
À 0.52[a]
À 0.9[a,d]
30
33
This work
This work
15,28
À 0.32[b,c]
À 0.86[e]
À 0.59[e]
(À 0.72;[f] À 0.74[g] and À 0.63[a,h,i]
)
2
À 0.92 (À 0.98)[j,k]
À 0.43
29
°
°
[a] At 25 C. [b] At 90 C. [c] pH 5.6, [Zn(II)]=10 mM. [d] pH 5.9, [Zn]=
°
°
°
10 mM. [e] pH=7.5, 40 C; [f] Conditions: pH 7.0, 25 C. [g] pH 6.5, 50 C. [h]
pH 6.6. [i] With high and low pKa, respectively. [j] Considering all substrates
°
examined and two βlg. [k] At 90 C and pH 6.5.
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
4
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100299
Chemistry—A European Journal
but, in this case, it survives long enough to allow isomerization to
occur. For this reason, it was suggested a strong interaction of the
transition state with the dinuclear catalyst leading to significant
dissipation of the developing negative charge in the reaction path
towards this intermediate. Catalyst 2 binds so strongly to the
phosphate to resemble a phosphate triester.[16,29,32] This facilitates
the intramolecular nucleophilic attack and is consistent with the
relatively rather negative βlg (À 0.92 for the less acidic leaving
groups). The picture for catalyst 1 is different: no intermediate is
formed, and the slope of the plot is less negative (À 0.59 under the
same conditions used for AuNP1). This implies a lower stabilization
of the phosphorane but also a significant contribution to leaving
group departure through proton transfer. AuNP1 presents
similarities with 1 and 2. On one side it does not stabilize the
phosphorane enough as is the case of catalyst 1. On the other
hand, its nucleophilic contribution to catalysis is important as for
catalyst 2. The better assistance in leaving group departure
exerted by catalyst 1 compared to AuNP1 results in an enhanced
catalytic performance by the nanoparticles as the leaving groups
become more acidic. The poorer stabilization of the phosphorane
exerted by the nanoparticles may be the result of the partial
detachment of the phosphate from the catalytic site required to
allow the reaction to occur as suggested by the thermodynamic
studies reported above. Thus, the presence of the nucleobase by
increasing the stabilization of the ground state, not only negatively
affects the efficiency of the catalytic process but also its
mechanism. Nevertheless, AuNP1 is still a better catalyst than 1.
Finally it should be pointed out that the behavior of all the above
catalysts is different from that of aqueous Zn(II) (at slightly acidic
pH to prevent precipitation).[33] Zn(II) catalysis provides in this case
an important general acid catalysis contribution, particularly for
the less acidic alcohols. The Brønsted plot shows βlg =À 0.32 and
À 0.9 for the less and more acidic leaving groups (alkyl and aryl
derivatives), respectively. Accordingly, alkyl alcohols depart as
neutral species while aryl derivatives depart as oxyanions. Notably,
the proposed mechanism requires only a single metal ion contrary
to catalysts 1, 2 and AuNP1 that are dinuclear ones.
Figure 4. Brønsted plot for the cleavage of substrates 3 by catalysts AuNP1
(blue symbols) and 1 (red symbols). Conditions: [AuNP1]=5.0×10À 6 M,
[1]=5.0×10À 5 M, [3]=3.0×10À 5 M, [HEPES]= 0.01 M, pH 7.5, 40 C. Note the
°
difference in concentration of the two catalysts ([1]=10×[AuNP1].
higher pKa of the leaving groups βlg =À 1.34 while with more
acidic ones βlg =À 0.52 (estimated value). The βlg for the different
catalysts and the base-catalyzed, background reaction are reported
in Table 3.
The mechanism of cleavage of uridine 3’-phosphate diesters
by AuNP1
Conclusion
The analysis of Table 3 reveals a quite diverse picture for the
cleavage of uridine 3’-phosphate diesters in the presence of
dinuclear Zn(II) catalysts. This diversity was already pointed out by
Mikkola et al. when analyzing a large collection of catalysts.[15] The
base-catalyzed reaction indicates a mechanism ((b) in Scheme 1)
in which an intermediate phosphorane is formed (analogously to
the tetrahedral intermediate in carboxylate esters hydrolysis).[31]
The break point of the Brønsted plot of the base-catalyzed process
occurs at pKa =12.58. This value is close to that reported for the 2’
hydroxyl of uridine 3’-phosphate ethyl ester (12.85). This similarity
is striking and resembles what one would expect for a quasi-
symmetrical reaction (attack by the nucleophile/leaving group
departure) despite being an intramolecular process. The base-
catalyzed process, however, does not allow isomerization of the
starting material to the 3’ derivative indicating the intermediate
phosphorane is short-lived under these conditions. For catalyst 2
also the formation of a phosphorane intermediate was indicated
We have compared AuNPs active in the cleavage of RNA model
substrates and operating with a dinuclear mechanism with other
two catalysts also based on a dinuclear catalytic site. The picture
that emerges is quite diverse for the three catalytic systems.
Catalyst 2 represents the extreme case of maximum stabilization
of the phosphorane (Scheme 1 (b)) that forms in the trans-
formation of the reagent into the cyclic product 2’,3’-cUMP. This
catalyst appears to be able to dissipate most of the charge
developed during the evolution to products. Assistance in leaving
group departure is possibly observed with this catalyst only for
the more acidic alcohols (pKa <12.4). On the contrary catalyst 1
does not stabilize enough the phosphorane but provides
significant assistance in leaving group departure. AuNP1 share
properties of both catalysts. On one side, as in the case of 1, the
phosphorane is not stabilized enough to become an intermediate.
On the other, as in the case of 2 for the less acidic alcohols, little
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
5
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100299
Chemistry—A European Journal
[8] C. Fasting, C. A. Schalley, M. Weber, O. Seitz, S. Hecht, B. Koksch, J.
Dernedde, C. Graf, E. Knapp, R. Haag, Angew. Chem. Int. Ed. 2012, 51,
10472–10498; Angew. Chem. 2012, 124, 10622–10498.
[9] P. Scrimin, M. Cardona, C. Prieto, L. J. Prins, Multivalency as a Design
Criterion in Catalyst Development, Ch. 7, Catalyst Development in
Multivalency: Concepts, Research & Applications, J. Huskens, L. J. Prins, R.
Haag, B. J. Ravoo, Eds., John Wiley & Sons Ltd., 2017.
[10] J. Czescik, S. Zamolo, T. Darbre, R. Rigo, C. Sissi, A. Pecina, L. Riccardi,
M. D. Vivo, F. Mancin, P. Scrimin, Angew. Chem. Int. Ed. 2021, 60, 1423–
1432; Angew. Chem. 2021, 133, 1443–1432.
assistance in leaving group departure is provided. The most
important contribution to catalysis is hence nucleophilic.
It is interesting to note that, despite that AuNP1 owe most of
its efficiency to a high affinity for the uridine 3’-phosphate diesters
(see Table 1), this is partly lost on the way to the transition state.
Thus, although in principle it should behave like catalyst 2, it fails
to stabilize enough the transient phosphorane. The uridine moiety,
while it is beneficial in stabilizing the ground state, is detrimental
to reach the optimum geometry in the transition state. Neither
catalyst 1 nor 2 appear to be able to interact with the uridine
moiety of the substrates, likely because their catalytic sites are
rather well defined and little flexible, contrary to that present on
the monolayer passivating the gold nanoparticles. One may argue
that better performing AuNPs could take advantage of the
presence of cationic groups to assist the metal ion in the catalytic
process as in the case of catalyst 2. We have observed that this is
indeed the case for very challenging substrates such as DNA.[10]
Our results show how very subtle changes, in both the catalyst
and substrate, may significantly affect the mechanism of metal ion
catalyzed cleavage of RNA phosphodiesters.
[11] F. Mancin, L. Prins, P. Pengo, L. Pasquato, P. Tecilla, P. Scrimin, Molecules
2016, 21, 1014.
[12] a) F. Manea, F. B. Houillon, L. Pasquato, P. Scrimin, Angew. Chem. Int. Ed.
2004, 43, 6165; Angew. Chem. 2004, 116, 6291–6169; b) H. Wei, E. Wang,
Chem. Soc. Rev. 2013, 42, 6060–6093; c) J. Wu, X. Wang, Q. Wang, Z. Lou,
S. Li, Y. Zhu, L. Qin, H. Wei, Chem. Soc. Rev. 2019, 48, 1004–1076; d) L.
Gabrielli, L. J. Prins, F. Rastrelli, F. Mancin, P. Scrimin, Eur. J. Org. Chem.
2020, 5044–5055.
[13] J. Czescik, S. Zamolo, T. Darbre, F. Mancin, P. Scrimin, Molecules 2019,
24, 2814.
[14] S. Mikkola, T. Lönnberg, H. Lönnberg, Beilstein J. Org. Chem. 2018, 14,
803–837.
[15] H. Korhonen, T. Koivusalo, S. Toivola, S. Mikkola, Org. Biomol. Chem.
2013, 11, 8324–8339.
[16] H. Korhonen, N. H. Williams, S. Mikkola, J. Phys. Org. Chem. 2013, 26,
182–186.
[17] a) S. Aoki, E. Kimura, Chem. Rev. 2004, 104, 769–788; b) E. Kimura, H.
Kitamura, K. Ohtani, T. Koike, J. Am. Chem. Soc. 2000, 122, 4668; c) M.
Shionoya, E. Kimura, M. Shiro, J. Am. Chem. Soc. 1993, 115, 6730.
[18] M. Diez-Castellnou, F. Mancin, P. Scrimin, J. Am. Chem. Soc. 2014, 136,
1158–61.
Acknowledgements
[19] G. Zaupa, C. Mora, R. Bonomi, L. J. Prins, P. Scrimin, Chem. Eur. J. 2011,
17, 4879–89.
Funding from the European Union Framework Programme for
Research and Innovation Horizon 2020 (H2020-MSCA-ITN-2016
MMBio – 721613) and Fondazione Cariparo (“Ricerca Scientifica di
Eccellenza” Grant “SELECT”) is gratefully acknowledged.
[20] L. Riccardi, L. Gabrielli, X. Sun, F. Biasi, F. Rastrelli, F. Mancin, M. De Vivo,
Chem 2017, 3, 92–109.
[21] G. Guarino, F. Rastrelli, P. Scrimin, F. Mancin, J. Am. Chem. Soc. 2012,
134, 7200–7203.
[22] a) Q. Wang, H. Lönnberg, J. Am. Chem. Soc. 2006, 128, 10716–10728;
b) Q. Wang, E. Leino, A. Jancsó, I. Szilágyi, T. Gajda, E. Hietamäki, H.
Lönnberg, ChemBioChem 2008, 9, 1739–1748; c) M. Laine, T. Lönnberg,
M. Helkearo, H. Lönnberg, Inorg. Chim. Acta 2016, 452, 111–117.
[23] Y. Lyu, L. Morillas-Becerril, F. Mancin, P. Scrimin, J. Hazard. Mater. 2021,
415, 125644.
[24] A. Yatsimirsky, Coord. Chem. Rev. 2005, 249, 1997–2011.
[25] R. Cacciapaglia, S. D. Stefano, L. Mandolini, Acc. Chem. Res. 2004, 37,
113–122.
Conflict of Interest
The authors declare no conflict of interest.
Keywords: Brønsted plot · gold nanoparticles · metal ion
catalysis · phosphate cleavage · RNA cleavage
[26] M.-Y. Yang, J. P. Richard, J. R. Morrow, Chem. Commun. 2003, 2832–
2833.
[27] O. Iranzo, A. Y. Kovalevsky, J. R. Morrow, J. P. Richard, J. Am. Chem. Soc.
2003, 125, 1988–1993.
[28] J. R. Morrow, T. L. Amyes, J. P. Richard, Acc. Chem. Res. 2008, 41, 539–48.
[29] H. Korhonen, S. Mikkola, N. H. Williams, Chem. Eur. J. 2012, 18, 659–670.
[30] H. Lönnberg, R. Strömberg, A. Williams, Org. Biomol. Chem. 2004, 2,
2165–2167.
[31] P. Scrimin, P. Tecilla, U. Tonellato, J. Org. Chem. 1994, 59,10–24.
[32] M. Kosonen, K. Hakala, H. Lönnberg, J. Chem. Soc. Perkin Trans. 2 1998,
663–670.
[1] a) D. Desbouis, I. Troitsky, M. Belousoff, L. Spiccia, B. Graham, Coord.
Chem. Rev. 2012, 256, 897–937; b) F. Mancin, P. Scrimin, P. Tecilla, Chem.
Commun. 2012, 48, 5545–5559; c) F. Mancin, P. Scrimin, P. Tecilla, U.
Tonellato, Chem. Commun. 2005, 2540–2548.
[2] R. Wolfenden, Annu. Rev. Biochem. 2011, 80, 645–667.
[3] W. Yang, Q. Rev. Biophys. 2011, 44, 1–93.
[33] S. Mikkola, E. Stenman, K. Nurmi, E. Yousefi-Salakdeh, R. Strömberg, H.
[4] G. Palermo, A. Cavalli, M. L. Klein, M. Alfonso-Prieto, M. Peraro, M.
De Vivo, Acc. Chem. Res. 2015, 48, 220–228.
Lönnberg, J. Chem. Soc. Perkin Trans. 2 1999, 1619–1626.
[5] M. Diez-Castellnou, A. Martinez, F. Mancin, Adv. Phys. Org. Chem. 2017,
51, 129–186.
[6] J. Chin, Acc. Chem. Res. 1991, 24, 145–152.
[7] D. Desbouis, I. Troitsky, M. Belousoff, L. Spiccia, B. Graham, Coord. Chem.
Rev. 2012, 256, 897–937.
Manuscript received: January 25, 2021
Accepted manuscript online: March 29, 2021
Version of record online: ■■■, ■■■■
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
6
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
Dr. J. Czescik, Prof. Dr. F. Mancin,
Prof. Dr. R. Strömberg*, Prof. Dr. P.
Scrimin*
1 – 7
The Mechanism of Cleavage of RNA
Phosphodiesters by a Gold Nanopar-
ticle Nanozyme
Uridine 3’-alkyl- and aryl-phospho-
diesters are easily cleaved by gold
nanozymes functionalized with Zn(II)
ligands with a nucleophilic
mechanism. The catalytic site requires
two metal ions for catalysis and at
least a third one for binding of the
uridine nucleobase.