Tuning the exchange dynamics of boronic acid hydrazones and oximes with pH and redox control

Article abstract of DOI:10.1039/d1ob00191d

Dynamic bonds continually form and dissociate at equilibrium. Carbonyl compounds with proximal boronic acids, including 2-formylphenylboronic acid (2-FPBA), have been reported to form highly dynamic covalent hydrazone and oxime bonds in physiological conditions, but strategies to tune the dynamics have not yet been reported. Here, we characterize the dynamics of 2-FPBA-derived hydrazones and oximes and account for both the rapid rate of formation (～10²-10³M^?1s^?1) and the relatively fast rate of hydrolysis (～10^?4s^?1) at physiological pH. We further show that these substrates undergo exchange with α-nucleophiles, which can be reversibly paused and restarted with pH control. Finally, we show that oxidation of the arylboronic acid effectively abolishes the rapid dynamics, which slows the forward reaction by more than 30?000 times and increases the hydrolytic half-life from 50 minutes to 6 months at physiological pH. These results set the stage to explore these linkages in dynamic combinatorial libraries, reversible bioconjugation, and self-healing materials.

Full text of DOI:10.1039/d1ob00191d

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Tuning the exchange dynamics of boronic acid
hydrazones and oximes with pH and redox
control†
Cite this: Org. Biomol. Chem., 2021,
19, 4986
Gun Su Han
and Dylan W. Domaille
*
Dynamic bonds continually form and dissociate at equilibrium. Carbonyl compounds with proximal
boronic acids, including 2-formylphenylboronic acid (2-FPBA), have been reported to form highly
dynamic covalent hydrazone and oxime bonds in physiological conditions, but strategies to tune the
dynamics have not yet been reported. Here, we characterize the dynamics of 2-FPBA-derived hydrazones
and oximes and account for both the rapid rate of formation (∼102–103 M−1 s−1) and the relatively fast rate
of hydrolysis (∼10−4 s−1) at physiological pH. We further show that these substrates undergo exchange
with α-nucleophiles, which can be reversibly paused and restarted with pH control. Finally, we show that
oxidation of the arylboronic acid effectively abolishes the rapid dynamics, which slows the forward reac-
tion by more than 30 000 times and increases the hydrolytic half-life from 50 minutes to 6 months at
physiological pH. These results set the stage to explore these linkages in dynamic combinatorial libraries,
reversible bioconjugation, and self-healing materials.
Received 1st February 2021,
Accepted 12th May 2021
DOI: 10.1039/d1ob00191d
rsc.li/obc
Introduction
Dynamic covalent bonds (DCBs) are reversible covalent bonds
at thermodynamic equilibrium.1–3 Two of the most commonly
used DCBs are hydrazones and oximes, formed from an alde-
hyde or ketone and
a
hydrazine or alkoxyamine,
respectively.4–11 As a result of their broad availability, ease of
incorporation into more complex (macro)molecules, and
ability to exchange, hydrazones and oximes have been used to
reversibly tag biomolecules12 and select for self-assembled
inhibitors in protein-directed dynamic combinatorial
chemistry.13
One of their biggest limitations in dynamic processes,
however, is the relatively slow rates of formation and exchange
in aqueous solvents.13 Aniline-based organocatalysts accelerate
the kinetics, but the need for a high catalyst concentration
(>50 mM) limits the broad use of this strategy, especially for
live-cell labeling applications.14,15 Neighboring group acti-
vation provides a complementary approach16–19 to increase the
rate of hydrazones and/or oximes formation.20–22 A variety of
functional groups proximal to the reacting carbonyl, including
o-pyridyl,16 o-phosphate,17 o-formyl,18 and o-boronic acids dra-
matically accelerate the rate of formation of hydrazones and/or
Fig. 1 2-FPBA-derived hydrazones and oximes are activated by the
proximal boronic acid and Brønsted acid catalysis. We hypothesized that
the dynamics could be reversibly paused and activated with pH control
and irreversibly slowed by oxidizing the boronic acid with H2O2.
oximes.20–22 Of these, o-boronic acids are especially active
neighboring groups, accelerating hydrazone and oxime for-
mations several orders of magnitude faster than hydrazone/
oxime formations with benzaldehyde.
Though the reaction between and α-nucleophiles and
o-boronic aryl aldehydes/ketones dates back more than sixty
years,23–26 only recently have these reactions been shown to
proceed exceptionally fast in physiological conditions, even at
micromolar concentrations.20,21,27,28 Despite their emerging
importance in site-specific bioorthogonal labels,29,30 oxida-
Department of Chemistry, Colorado School of Mines, Golden, CO, USA.
E-mail: ddomaille@mines.edu
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
d1ob00191d
4986 | Org. Biomol. Chem., 2021, 19, 4986–4991
This journal is © The Royal Society of Chemistry 2021

View Article Online
Organic & Biomolecular Chemistry
Paper
tively activated metal chelators,31–33 and potential for use in
dynamic combinatorial libraries, a comprehensive characteriz-
ation of the reversible dynamics of hydrazones and oximes
with neighboring boronic acids, susceptibility to exchange,
and responsivity to common molecular inputs has not been
reported.
Here, we characterize highly dynamic 2-formylphenylboro-
nic acid (2-FPBA)-derived hydrazone and oxime bonds by
accounting for both the rapid rate of formation (∼102–103 M−1
s−1) and the relatively fast rate of hydrolysis (t1/2 ∼50 minutes)
at physiological pH. We demonstrate that the resulting hydra-
zone/diazaborine or oxime can exchange with another
α-nucleophile and that the exchange can be temporarily
paused and restarted with pH control. Finally, we highlight
that treatment with H2O2 oxidizes the arylboronic acid catalyst
to a far less active hydroxyl group, effectively abolishing the
rapid dynamics (Fig. 1). Taken together, our studies demon-
strate that the rapid dynamics of 2-FPBA-derived hydrazones
and oximes can be reversibly sped up and slowed down and
irreversibly abolished with simple molecular inputs. These
results provide a powerful way to tune the exchange dynamics
of this class of dynamic bonds, setting the stage to explore
their utility in reversible bioconjugation reactions and
dynamic combinatorial library generation.
Fig. 2 2-FPBA forms highly dynamic hydrazone and oxime bonds. (A)
The reaction kinetics were measured for 2-FPBA and AHz and for
2-FPBA and MHA. (B) Formation of 1a was monitored as a function of
time. (C) Formation of 2a was monitored as a function of time. The red
line represents a fit to the kinetic model. Kinetic traces were repeated in
triplicate; rate constants are reported as the average of three runs with
independent fitting to the model SD.
second-order model fits the data poorly (Fig. S2†). We hypoth-
esized that proximal boronic acid was also accelerating the rate
Results and discussion
2-Formylphenylboronic acid (2-FPBA) has previously been of hydrolysis and fitted the data to a reversible model that is
shown to form a hydrazone or oxime at an exceptionally fast second-order in the forward direction (k1) and first-order in
rate. The rate acceleration has been attributed to the intra- hydrolysis in the back reaction (k−1).4 This provides an excel-
molecular catalytic effect of o-boronic acid.20 To confirm the lent fit to the kinetic data.
proximity effect of the boronic acid, we measured the rate of
hydrazone and oxime formations from 2-FPBA and its para- = 148 4 M−1 s−1) with concomitant rates of hydrolysis six
isomer, 4-formylphenylboronic acid (4-FPBA) with acethydra- orders of magnitude slower (k−1 = 2.3 0.1 × 10−4
−1; t1/2
The reaction between 2-FPBA and AHz proceeds rapidly (k1
s
zide (AHz) and O-methylhydroxylamine (MHA) as model hydra- ∼50 minutes) at 15 °C (Fig. 2B). At pD 7.5, we observed only
zide and alkoxyamine nucleophiles, respectively. 4-FPBA was DAB 1a (Fig. S3†). The reaction between 2-FPBA and MHA
combined with equimolar AHz or MHA (1.85 mM) in phos- forms 2a quickly and hydrolyzes at a relatively fast rate (k1
=
phate buffer (PB, 200 mM, pD 7.5) at room temperature, and 460 14 M−1 s−1 and k−1 = 3.2 0.2 × 10−4
s
−1, respectively)
product formation was monitored with time-course NMR spec- (Fig. 2C). The dramatic rate difference between hydrazone or
troscopy. The reaction between 4-FPBA and MHA proceeds oxime formation from 2-FPBA compared to 4-FPBA supports
with a rate constant of 3.3 × 10−2 M−1 s−1, while the reaction the idea that the rapid rate acceleration originates from a
between 4-FPBA and AHz is roughly 10-fold slower (k1 = 3.8 × neighboring group effect of the boronic acid. It is also impor-
10−3 M−1 s−1) (Fig. S1†).
tant to note that the presence of the o-boronic acid accelerates
When we repeated the experiment with the ortho-isomer the hydrolysis of the diazaborine and oxime, and the back rate
2-FPBA, the product formation was too fast to monitor with of hydrolysis must be accounted for when measuring the kine-
NMR spectroscopy. Instead, time-course ultraviolet-visible tics of this class of compounds.
spectroscopy (UV-vis) was used to track product formation.
Next, we completed the same kinetic analysis at 35 °C,
2-FPBA (100 μM) was combined with equimolar AHz or MHA 55 °C, and 75 °C for the formations of 1a and 2a. Eyring ana-
(100 μM) in phosphate buffer (PB, 100 mM, pD 7.5) at 15 °C, lysis yielded activation parameters of ΔH‡ = 13.3 kJ mol−1 and
and product formation was monitored with UV-visible spec- ΔS‡ = −157 J mol−1 to form 1a and ΔH‡ = 59.0 kJ mol−1 and
troscopy. As reported by the Bane group, the reaction of ΔS‡ = −109 J mol−1 to re-form 2-FPBA and AHz (Fig. S4A and
2-FPBA and AHz initially forms a hydrazone, which, in B†). At room temperature, this corresponds to a low energy
aqueous solution, rapidly equilibrates with its cyclized barrier to formation (ΔG‡ = 60 kJ mol−1) with a higher energy
product, diazaborine (DAB) (Fig. 2A, 1a).21,27 The reaction of barrier to hydrolysis (ΔG‡ = 91 kJ mol−1). Oxime 2a has acti-
2-FPBA and MHA yields oxime 2a (Fig. 2A, 2a). An irreversible vation parameters of ΔH‡ = 18.6 kJ mol−1 and ΔS‡ = −129 J K−1
This journal is © The Royal Society of Chemistry 2021
Org. Biomol. Chem., 2021, 19, 4986–4991 | 4987

View Article Online
Paper
Organic & Biomolecular Chemistry
mol−1 to form 2a and ΔH‡ = 57.4 kJ mol−1 and ΔS‡ = −111 J (iBHz), and tert-butylcarbazate (tBCz)), as well as an electron-
K−1 mol−1 to re-form the constituent 2-FPBA and MHA build- deficient hydrazide (cyanoacetohydrazide (CAHz)).
ing blocks. At room temperature, oxime formation proceeds
with a low energy barrier of ΔG‡ = 57 kJ mol−1 and a higher dependent on their steric effects (Fig. 4A). While PivHz shows
barrier of ΔG‡ = 91 kJ mol−1 to hydrolysis (Fig. S4C and D†).
no exchange with DAB 1a (Fig. 4A, ( ); Fig S5a and b†) and
The hydrazides show different extents of exchanges that are
Hydrazone and oxime formations are well-established to be iBHz shows limited exchange (Fig. 4A ( ); Fig. S6a and b†),
dependent on pH conditions.27,34 To explore how sensitive the tBCz, which has its tert-butyl group one atom farther from the
dynamics of this particular class of 2-FPBA-derived hydra- hydrazide compared to pivalic acid hydrazide, exchanges and
zones/diazaborines and oximes are to pH, we measured the equilibrates to 92% 1e after 10 hours (Fig. 4A ( ); Fig. S7a and
kinetics of formation and hydrolysis of 1a and 2a at 15 °C at b†). CAHz also exchanges with 1a and equilibrates to 84% 1d,
pH 6.5, 7.4, 8.5, 9.5, and 12.0. Because our kinetic analysis despite the inclusion of a strong electron-withdrawing group
enables the extraction of both k1 and k−1, we can delineate the (Fig. 4A ( ); Fig S8a and b†). Repeating the same experiments
influence of pH on each process independently. Within the at pD 12.0 led to no measurable exchange after 15 hours for all
explored range of pH, k1 and k−1 decrease as a function of pH, hydrazides (Fig. S9–S12†).
resulting in three orders of magnitude difference in the rates
between pH 6.5 and 12.0 (Fig. 3A and B). The Keq for DAB 1a of
reaches a maximum at pH 8.5, and then decreases slightly at O-ethylhydroxylamine
We next investigated how oxime 2a exchanges with a range
increasingly bulky alkoxyamines, including
(EHA), O-isopropylhydroxylamine
higher pH conditions (Fig. 3C). We attribute the lowest Keq at (iPHA), and O-tert-butylhydroxylamine (t-BHA). All alkoxy-
pH 6.5 to increased formation of acyl hydrazone, which is amines, including the very bulky t-BHA, exchange (Fig. 4B; ESI
hydrolytically less stable than DAB 1a. The mechanistic origin Fig. S13a–S15b†). There is a slight increase in exchange as the
of why the Keq of DAB 1a reaches a maximum at pH 8.5 is not size of the alkyl group increases; however, steric effects do not
immediately apparent. The Keq for oxime 2a trends downward affect the exchange to the same degree as they do in the hydra-
as the pH increases (Fig. 3D). From pH 6.5–12, DAB 1a and zide examples. We attribute the difference in the exchange
oxime 2a show Keq values on the order of 105–106 despite large behaviors between the diazaborines and oximes to the absence
changes in their k1 and k−1 values.
of the hetereocyclic ring in case of the oximes. Similar to the
Accessing facile exchange chemistry is a central component results we observed with hydrazone exchange, oxime
to identifying reactions for dynamic combinatorial chemistry exchanges were significantly slower at pD 12.0. Less than 10%
and reversible bioconjugation. Because k1 and k−1 are sensitive exchange was observed after 14 hours for each alkoxyamine
to pH, we reasoned that the rate of hydrazone exchange of 1a (Fig. S16–S18†).
or oxime exchange of 2a could be controlled by simply altering
Next, we turned to using pD as an easily reversible switch to
the pH of the solution. To explore this idea, we investigated pause and restart the exchange process. When 1a was formed
how 1a exchanges with a competing hydrazide. 1a was formed in situ from AHz (3 equiv.) at pD 7.5 and treated with tBCz (0.5
in situ in pD 7.5 in buffered D2O (200 mM PB). We surveyed a
range of competing hydrazides, including bulky hydrazides
(pivalic acid hydrazide (PivHz), isobutyric acid hydrazide
Fig. 4 Exchange of 1a with a range of acylhydrazides and exchange of
2a with a range of alkoxyamines. (A) 1a was formed in situ from 2-FPBA
and AHz (3 equiv.) in aqueous buffer (200 mM PB, pD 7.5) and sub-
Fig. 3 Effect of pH on the rate of formation and hydrolysis of 1a and
2a. (A) k1 rates of 1a (filled squares) and oxime 2a (empty squares) as a
function of pH. (B) k−1 rates of 1a (filled squares) and 2a (open squares).
(C) Keq values 1a as a function of pH. (D) Keq values for oxime 2a. Values
were measured in triplicate. Error bars represent S.D. of three indepen-
dent measurements.
sequently treated with either iBHz (3 equiv.) to give 1b ( ), PivHz (3
equiv.) to give 1c ( ), CAHz (3 equiv.) to give 1d ( ), or tBCZ (3 equiv.) to
give 1e ( ). (B) 2a was formed in situ from 2-FPBA and MHA (3 equiv.) in
aqueous buffer (200 mM PB, pD 7.5) and subsequently treated with
either EHA (3equiv.) to form 2b ( ), iPHA (3 equiv.) to form 2c ( ), and
t-BHA (3 equiv.) to form 2d ( ).
4988 | Org. Biomol. Chem., 2021, 19, 4986–4991
This journal is © The Royal Society of Chemistry 2021

View Article Online
Organic & Biomolecular Chemistry
Paper
equiv.), the ratio of 1a : 1e equilibrates to 57 : 43 (Fig. 5A).
However, when the pD was adjusted to 12.0 before adding
tBCz (0.5 equiv.), no exchange was observed for 15 hours
(Fig. 5B). The difference in the exchange behaviors inspired us
to explore subsequent exchanges with pD control. 1a was
formed in situ from AHz (3 equiv.) and treated with tBCz (0.5
equiv.) at pD 7.5 in buffered D2O. After the initial exchange
was complete, 2.5 equivalents of tBCz were added, and the
exchange was monitored with time-course NMR spectroscopy
(Fig. 5C). The subsequent exchange finished in 6 hours with
88% 1e. However, when we adjusted the pD to 12 prior the
addition of 2.5 equivalents of tBCz, no subsequent exchange
was observed. When the pD was readjusted to 7.5, the
exchange immediately resumed; after 7 hours, the mixture
equilibrated to 88% 1e again (Fig. 5D).
In the case of oxime exchange, methyloxime 2a and ethylox-
ime 2b are of similar thermodynamic stability (Fig. 6A) and
undergo pH-dependent exchange, which significantly slows
down at pD 12.0 (Fig. 6B). We thus reasoned that the amount
of 2a or 2b can be controlled with off-stoichiometric control of
the nucleophile. Oxime 2a was formed in situ from MHA (3
equiv.) and treated with EHA (3 equiv.) at pD 7.5 in buffered
D2O. After the initial exchange was complete with ∼50% of
each oxime, 24 equivalents of MHA were added, and the
exchange was monitored with time-course NMR spectroscopy
(Fig. 6C). The subsequent exchange finished in 20 hours with
Fig. 6 Pausing and restarting the exchange of 2-FPBA oximes. (A) 2a
was formed in situ from 2-FPBA and MHA (3 equiv.) in aqueous buffer
(200 mM PB, pD 7.5) and subsequently treated with EHA (3 equiv.) to
equilibrate with 2b at pD 7.5. (B) Under the same conditions as (A) but at
pD 12.0, the exchange does not occur. (C) Equilibrating between 2a and
2b as in (A) but followed with an additional aliquot of EHA (24 equiv.,
dashed red line) drives the equilibrium to a 2a : 2b ratio of 18 : 82. (D)
Establishing an equilibrium between 2a and 2b as in (A) followed by a pD
adjustment to 12.0 (dashed blue line) and addition of another aliquot of
EHA (24 equiv., dashed red line) does not initiate exchange until the pD
is adjusted back to 7.5 (dashed green line).
82% 2a. However, when we adjusted the pD to 12 prior the
addition of 24 equivalents of MHA, the mixture showed no
subsequent exchange. When the pD was readjusted to 7.5, the
exchange immediately resumed; after 20 hours, the mixture
equilibrated to 83% oxime 2a (Fig. 6D). Taken together, these
experiments demonstrate that pH control can pause and reac-
tivate exchange dynamics of 2-FPBA-derived DAB and oximes.
Next, we explored strategies to irreversibly abolish the rapid
dynamics with an orthogonal input to pH control. Arylboronic
acids undergo H2O2-mediated oxidation to their corresponding
phenols.35,36 Because the rapid dynamics of 2-FPBA-derived
hydrazones and oximes results from activation of the carbonyl
by the proximal boronic acid group, we recognized that oxi-
dation would effectively replace the boronic acid with a
hydroxyl group. We thus hypothesized that treatment of DAB
1a and oxime 2a with H2O2 would serve as a mild molecular
input to abolish the rapid exchange dynamics.
Fig. 5 Pausing and restarting the exchange of 2-FPBA diazaborines. (A)
1a was formed in situ from 2-FPBA and AHz (3 equiv.) in aqueous buffer
(200 mM PB, pD 7.5) and subsequently treated with tBCz (0.5 equiv.,
tBCz) to equilibrate with 1e at pD 7.5. (B) If the pD is adjusted to 12
before adding tBCz (0.5 equiv.), the exchange does not occur. (C) 1a and
In support of this hypothesis, we first measured the k1 and
k−1 for hydrazone and oxime formation from salicylaldehyde
(Sal). The reaction of Sal and AHz yields acyl hydrazone 3 (ESI
Fig. S19†), which forms with a rate constant of 2.6 × 10−3 M−1
1e were initially equilibrated as in (A), and the resulting mixture was s−1 and hydrolyzes with a rate constant of 4.8 × 10−7 s−1. Thus,
treated with an additional aliquot of tBCz (2.5 equiv., dashed red line).
The resulting mixture equilibrated to a 1a : 1e ratio of 12 : 88. (D) When
the pD was adjusted to 12.0 (dashed blue line) before adding the second
aliquot of tBCz (2.5 equiv., dashed red line), no subsequent exchange
hydrazone 3, which is derived from salicylaldehyde, forms 5
orders of magnitude more slowly and hydrolyzes 3-orders of
magnitude more slowly than 2-FPBA-derived DAB 1a. The reac-
tion of Sal and MHA yields oxime 4a (Fig. 7A). 4a, derived from
salicylaldehyde, forms and hydrolyzes more slowly (k1 = 1.4 ×
occurred for 15 hours. Readjusting the pD to 7.5 (dashed green line)
immediately resumed the subsequent exchange.
This journal is © The Royal Society of Chemistry 2021
Org. Biomol. Chem., 2021, 19, 4986–4991 | 4989

View Article Online
Paper
Organic & Biomolecular Chemistry
each hydrazide (Fig. S23–S26†) or between 4a and each alkoxy-
amine (Fig. 8A and Fig. S27–S29†).
Next, we reasoned that a target oxime population could be
established by using the 2-FPBA and then ‘cementing’ it with
oxidation. Oxime 2a was formed in situ with MHA (3 equiv.)
and treated with EHA (3 equiv.) at pD 7.5 in buffered D2O.
When the mixture was oxidized with H2O2 (3 equiv.) and then
treated with 24 equivalents of MHA, no subsequent exchange
was observed over the course of 18 hours (Fig. 8B). Thus, in
contrast to pH inputs that enable reversible pausing and acti-
vation of the exchange dynamics, oxidation irreversibly
abolishes the rapid dynamics of 2-FPBA-derived oximes, even
at physiological pH.
Fig. 7 The dynamics of an oxime with an ortho hydroxy is slower than
that of an oxime with an ortho boronic acid. (A) The reaction of salicylal-
dehyde (1.85 mM) with MHA (1.85 mM) at pD 7.5 forms 4a 5-orders of
magnitude slower and hydrolyzes 3-orders of magnitude slower than
2-FPBA-derived 2a. (B) Conversion of the 2a to 4a with H2O2 (3 equiv.) is
complete in about 1 h.
Conclusions
To close, we have presented a kinetic characterization of
2-FPBA-derived hydrazones and oximes that accounts for both
their rapid formation and relatively fast hydrolysis. This
kinetic analysis reveals that the resulting diazaborines and
oximes have Keq values on the order of 105–106 in the pH
range 6.5–12.0. We further demonstrate that small alkyl hydra-
zides undergo facile exchange with DAB 1a, even when the
competing hydrazide has a strong electron withdrawing group.
Sterically congested hydrazides, however, show little to no
exchange. Alkoxyamines in exchange with oxime 2a are com-
paratively unaffected by steric effects. The rate of exchange can
be temporarily paused by raising the pH and restarted by
returning to physiological pH. Finally, we demonstrate that the
rapid dynamics can be irreversibly abolished by oxidizing the
proximal boronic acid to a phenol, which removes the neigh-
boring group catalysis. Taken together, these results enable
control over the formation kinetics and exchange dynamics
and offer new avenues for reversible bioconjugation and
dynamic combinatorial library generation.
10−2 M−1 s−1; k−1 = 4.5 × 10−8 s−1) than 2-FPBA-derived 2a.
Thus, replacing the boronic acid with a phenol slows the rate
of formation by more than 34 000 times and slows the rate of
hydrolysis by more than 8000 times. In contrast to oxime 2a,
which has a hydrolytic half-life of ca. 50 minutes at 15 °C at
physiological pH, the oxidized product has a hydrolytic half-
life of almost 6 months. Thus, oxidation of 1a and 2a into 3
and 4a, respectively, should effectively abolish the rapid
dynamics of the bond with oxidation.
Oxime 2a oxidizes cleanly to 4a with H2O2 as revealed by
time-course NMR spectroscopy (Fig. 7B and Fig. S20a, S21b†).
A near quantitative oxidation was achieved within 1 hour. DAB
1a also undergoes H2O2-mediated conversion to its corres-
ponding phenol 3 (Fig. S22†). In contrast to the oxidation of
oxime 2a to its phenol, which is complete in 1 hour, the oxi-
dation of 1a takes more than 2 days at room temperature.37
Reversibility assays reveal that 3 is resistant to exchange. After
15 hours, no observable exchange was observed between 3 and
Author contributions
G.H. and D.W.D. conceptualized the research. G.H. collected
the data. G.H. and D.W.D analyzed the data, and G.H. and D.
W.D. wrote the manuscript.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
Fig. 8 Unlike oxime 2a, oxime 4a does not undergo exchange at pD 7.5
(A) Oxime 4a does not exchange with EHA (3 equiv.) at pD 7.5. (B)
Treating 4a with EHA (3 equiv.) establishes an equilibrium between 4a
and 4b after 12 h. After H2O2-mediated oxidation (3 equiv., dashed blue
line), the addition of additional EHA (24 equiv., dashed red line) does not
initiate exchange.
This work was supported by start-up funds from the Colorado
School of Mines and the Colorado Office of Economic
Development and International Trade Advanced Industries
Proof-of-Concept grant. We are indebted to Dr Thomas Lee,
the director of the Central Analytical Laboratory and Mass
4990 | Org. Biomol. Chem., 2021, 19, 4986–4991
This journal is © The Royal Society of Chemistry 2021

View Article Online
Organic & Biomolecular Chemistry
Paper
Spectrometry Facility at the University of Colorado, Boulder for 18 P. Schmidt, L. Zhou, K. Tishinov, K. Zimmermann and
helping with the mass analysis of the compounds. The pur-
chase of Waters Synapt G2 HDMS was made possible by
National Institute of Health grant S10-RR026641.
D. Gillingham, Angew. Chem., Int. Ed., 2014, 53, 10928–
10931.
19 P. L. Higgs, A. J. Ruiz-Sanchez, M. Dalmina, B. R. Horrocks,
A. G. Leach and D. A. Fulton, Org. Biomol. Chem., 2019, 17,
3218–3224.
20 P. Schmidt, C. Stress and D. Gillingham, Chem. Sci., 2015,
6, 3329–3333.
Notes and references
21 O. Dilek, Z. Lei, K. Mukherjee and S. Bane, Chem.
Commun., 2015, 51, 16992–16995.
22 A. Bandyopadhyay and J. Gao, Chem. – Eur. J., 2015, 21,
14748–14752.
23 M. J. S. Dewar and R. C. Dougherty, J. Am. Chem. Soc., 1964,
86, 433–436.
1 Y. Jin, C. Yu, R. J. Denman and W. Zhang, Chem. Soc. Rev.,
2013, 42, 6634.
2 Y. Jin, Q. Wang, P. Taynton and W. Zhang, Acc. Chem. Res.,
2014, 47, 1575–1586.
3 S. Ulrich, Acc. Chem. Res., 2019, 52, 510–519.
4 A. Dirksen, S. Dirksen, T. M. Hackeng and P. E. Dawson,
J. Am. Chem. Soc., 2006, 128, 15602–15603.
5 R. Nguyen and I. Huc, Chem. Commun., 2003, 942.
6 L. Shen, N. Cao, L. Tong, X. Zhang, G. Wu, T. Jiao, Q. Yin,
J. Zhu, Y. Pan and H. Li, Angew. Chem., 2018, 130, 16724–
16728.
7 B. Levrand, W. Fieber, J.-M. Lehn and A. Herrmann, Helv.
Chim. Acta, 2007, 90, 2281–2314.
8 B. Levrand, Y. Ruff, J.-M. Lehn and A. Herrmann, Chem.
Commun., 2006, 2965–2967.
9 K. Kim, H. J. Cho, J. Lee, S. Ha, S. G. Song, S. Kim,
W. S. Yun, S. K. Kim, J. Huh and C. Song, Macromolecules,
2018, 51, 8278–8285.
10 M. N. Chaur, D. Collado and J.-M. Lehn, Chem. – Eur. J.,
2011, 17, 248–258.
11 S. R. Beeren, M. Pittelkow and J. K. M. Sanders, Chem.
Commun., 2011, 47, 7359.
12 S. Pfander and R. Fiammengo, Nucleic Acids Res., 2007, 35,
8.
24 M. J. S. Dewar and R. C. Dougherty, J. Am. Chem. Soc., 1962,
84, 2648–2649.
25 P. Tschampel and H. R. Snyder, J. Org. Chem., 1964, 29,
2168–2172.
26 H. R. Snyder, A. J. Reedy and Wm. J. Lennarz, J. Am. Chem.
Soc., 1958, 80, 835–838.
27 H. Gu, T. I. Chio, Z. Lei, R. J. Staples, J. S. Hirschi and
S. Bane, Org. Biomol. Chem., 2017, 15, 7543–7548.
28 C. J. Stress, P. J. Schmidt and D. G. Gillingham, Org.
Biomol. Chem., 2016, 14, 5529–5533.
29 T. I. Chio, H. Gu, K. Mukherjee, L. N. Tumey and
S. L. Bane, Bioconjugate Chem., 2019, 30, 1554–1564.
30 S. Cambray and J. Gao, Acc. Chem. Res., 2018, 51, 2198–
2206.
31 L. K. Charkoudian, D. M. Pham and K. J. Franz, J. Am.
Chem. Soc., 2006, 128, 12424–12425.
32 L. K. Charkoudian, D. M. Pham, A. M. Kwon,
A. D. Vangeloff and K. J. Franz, Dalton Trans., 2007, 5031.
33 L. K. Charkoudian, T. Dentchev, N. Lukinova, N. Wolkow,
J. L. Dunaief and K. J. Franz, J. Inorg. Biochem., 2008, 102,
2130–2135.
34 Q. Wang and K. J. Franz, Bioorg. Med. Chem. Lett., 2017, 27,
4165–4170.
35 H. G. Kuivila, J. Am. Chem. Soc., 1954, 76, 870–874.
36 H. G. Kuivila and K. A. Armour, J. Am. Chem. Soc., 1957, 79,
5659–5662.
37 The origin of the substantial rate difference between the
oxidation of a 2-FPBA-derived diazaborines and oximes is
currently under investigation in our laboratory and will be
reported shortly.
13 A. Canal-Martín and R. Pérez-Fernández, ACS Omega, 2020,
5, 26307–26315.
14 D. Larsen, A. M. Kietrys, S. A. Clark, H. S. Park, A. Ekebergh
and E. T. Kool, Chem. Sci., 2018, 9, 5252–5259.
15 V. T. Bhat, A. M. Caniard, T. Luksch, R. Brenk,
D. J. Campopiano and M. F. Greaney, Nat. Chem., 2010, 2,
490–497.
16 E. T. Kool, D.-H. Park and P. Crisalli, J. Am. Chem. Soc.,
2013, 135, 17663–17666.
17 O. Dilek, A. Sorrentino and S. Bane, Synlett, 2016, 27, 1335–
1338.
This journal is © The Royal Society of Chemistry 2021
Org. Biomol. Chem., 2021, 19, 4986–4991 | 4991