Selective Binding and Quantitation of Calcium with a Cobalt-Based Magnetic Resonance Probe

Article abstract of DOI:10.1021/jacs.9b02661

We report a cobalt-based paramagnetic chemical exchange saturation transfer (PARACEST) magnetic resonance (MR) probe that is able to selectively bind and quantitate the concentration of Ca²⁺ ions under physiological conditions. The parent LCo complex features CEST-active carboxamide groups and an uncoordinated crown ether moiety in close proximity to a high-spin pseudo-octahedral Co^II center. Addition of Na⁺, Mg²⁺, K⁺, and Ca²⁺ leads to binding of these metal ions within the crown ether. Single-crystal X-ray diffraction and solid-state magnetic measurements reveal the presence of a cation-specific coordination environment and magnetic anisotropy of Co^II, with axial zero-field splitting parameters for the Na⁺- A nd Ca²⁺-bound complexes differing by over 90%. Owing to these differences, solution-based measurements under physiological conditions indicate reversible binding of Na⁺ and Ca²⁺ to give well-separated CEST peaks at 69 and 80 ppm for [LCoNa]⁺ and [LCoCa]²⁺, respectively. Dissociation constants for different cation-bound complexes of LCo, as determined by ¹H NMR spectroscopy, demonstrate high selectivity toward Ca²⁺. This finding, in conjunction with the large excess of Na⁺ in physiological environments, minimizes interference from related cations, such as Mg²⁺ and K⁺. Finally, variable-[Ca²⁺] CEST spectra establish the ratio between the CEST peak intensities for the Ca²⁺- A nd Na⁺-bound probes (CEST_{80 ppm}/CEST_{69 ppm}) as a measure of [Ca²⁺], providing the first example of a ratiometric quantitation of Ca²⁺ concentration using PARACEST. Taken together, these results demonstrate the ability of transition metal PARACEST probes to afford a concentration-independent measure of [Ca²⁺] and provide a new approach for designing MR probes for cation sensing.

Full text of DOI:10.1021/jacs.9b02661

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC
Article
Selective Binding and Quantitation of Calcium
with a Cobalt-Based Magnetic Resonance Probe
Kang Du, Agnes E. Thorarinsdottir, and T. David Harris
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b02661 • Publication Date (Web): 04 Apr 2019
Downloaded from http://pubs.acs.org on April 4, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Selective Binding and Quantitation of Calcium with a Cobalt-Based
Magnetic Resonance Probe
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kang Du, Agnes E. Thorarinsdottir, and T. David Harris*
Department of Chemistry, Northwestern University, Evanston, Illinois 60208-3313, United States
ABSTRACT: We report a cobalt-based paramagnetic chemical exchange saturation transfer (PARACEST) magnetic resonance (MR) probe
that is able to selectively bind and quantitate the concentration of Ca2+ ions under physiological conditions. The parent LCo complex features
CEST-active carboxamide groups and an uncoordinated crown ether moiety in close proximity to a high-spin pseudo-octahedral CoII center.
Addition of Na+, Mg2+, K+, and Ca2+ leads to binding of these metal ions within the crown ether. Single-crystal X-ray diffraction and solid-
state magnetic measurements reveal the presence of a cation-specific coordination environment and magnetic anisotropy of CoII, with axial
zero-field splitting parameters for the Na+- and Ca2+-bound complexes differing by over 90%. Owing to these differences, solution-based
measurements under physiological conditions indicate reversible binding of Na+ and Ca2+ to give well-separated CEST peaks at 69 and 80
ppm for [LCoNa]+ and [LCoCa]2+, respectively. Dissociation constants for different cation-bound complexes of LCo, as determined by 1H
NMR spectroscopy, demonstrate high selectivity toward Ca2+. This finding, in conjunction with the large excess of Na+ in physiological
environments, minimizes interference from related cations, such as Mg2+ and K+. Finally, variable-[Ca2+] CEST spectra establish the ratio
between the CEST peak intensities for the Ca2+- and Na+-bound probes (CEST80 ppm/CEST69 ppm) as a measure of [Ca2+], providing the first
example of a ratiometric quantitation of Ca2+ concentration using PARACEST. Taken together, these results demonstrate the ability of
transition metal PARACEST probes to afford a concentration-independent measure of [Ca2+], and provide a new approach for designing MR
probes for cation sensing.
environment,10 pH,9e,11 temperature,12 and Zn2+ ions.13 The ex-
INTRODUCTION
changeable proton resonances, commonly from coordinating H2O,
carboxamides, and nitrogen heterocycles, are paramagnetically
shifted, thus minimizing interference from labile protons in biolog-
ical systems. Furthermore, the large chemical shifts allow for faster
proton exchange, hence more pronounced contrast can be realized.
The frequency-specific contrast afforded by PARACEST probes
enables simultaneous detection of more than one CEST peak. As a
result, the intensity ratio of two distinct CEST peaks that exhibit
different responses can provide an effective and concentration-in-
dependent measure of biomarkers.8b,9e,10c,11a,b,d–f,h,j,k
An ideal Ca2+-responsive PARACEST probe should feature a
recognition moiety that is moderately selective for Ca2+,8e yet can
reversibly bind other cations of concentrations that are relatively
constant in serum, in order to enable a ratiometric measurement.
One such cation is Na+, which exhibits a relatively constant con-
centration of ca. 140 mM in serum.14 In addition, the frequencies
of CEST peaks for Ca2+- and Na+-bound complexes should be well
separated to avoid interference, analogously to the attributes of a
19F probe.15 Along these lines, alkali and alkaline earth cations have
been shown to significantly influence the magnetic anisotropy of a
nearby paramagnetic metal ion, by causing distortions in the local
coordination environment.16 Because the proton hyperfine shift is
highly sensitive to changes in metal coordination environment and
magnetic anisotropy,17 the CEST peak frequency can be indicative
of the identity of the bound cation. As such, we set out to design a
probe that features (1) a cation binding moiety with proper affini-
ties toward Ca2+ and Na+ to allow for an equilibrium between the
Ca2+- and Na+-bound probes under physiological conditions, (2) a
paramagnetic center with magnetic properties and coordination en-
vironment that is highly sensitive toward the identities of cations
within its vicinity, and (3) a functional group capable of producing
CEST effects.
The concentration of Ca2+ ions in blood serum is a vital bi-
omarker for bone-related diseases, such as cancer,1 hyperparathy-
roidism,2 and Paget’s disease.3 These diseases are associated with
the dissolution of bone tissue, which releases Ca2+ into the blood
stream and results in hypercalcemia, a medical condition where the
total Ca2+ concentration in serum exceeds 2.6 mM.4 In current clin-
ical settings, the presence and extent of hypercalcemia is evaluated
by blood tests. This form of analysis provides only an estimate of
the total Ca2+ concentration in serum, with no information on the
spatial distribution or local concentration of Ca2+ near the bone le-
sion. As such, while blood tests can conveniently confirm the pres-
ence of hypercalcemia, they do not enable an assessment of the un-
derlying source and cause of high Ca2+ concentrations.5 For these
reasons, realization of an imaging technique able to quantitate the
local Ca2+ concentration near bone tissue would be highly useful in
the early detection of bone-related diseases and in pathological
studies.
Magnetic resonance imaging (MRI) is a non-invasive technique
that is particularly well suited for measuring the concentration of
Ca2+ near bone tissue owing to its unlimited depth penetration of
tissue and its ability to provide spatiotemporal images.6 Toward
this end, several GdIII-based probes have been developed to detect
Ca2+ ions by virtue of relaxivity changes upon binding Ca2+.7 Here,
extensive synthetic modifications have been employed to impart
selective binding of Ca2+ in the presence of other cations.7 Never-
theless, the utility of these probes is limited by heterogeneous bio-
distribution of Ca2+ and/or the probes themselves. It is therefore
critical to develop MRI probes capable of selectively binding and
quantitating Ca2+ through a concentration-independent method.
Lanthanide-8 and transition metal-based9 paramagnetic chemical
exchange saturation transfer (PARACEST) probes, which deliver
magnetization to bulk H2O through chemical exchange of protons,
have been reported to detect a number of biomarkers, such as redox
A high-magnetic anisotropy CoII complex that features coordi-
nated carboxamide ligands and a proximate crown ether satisfies
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
all of these criteria. Encouragingly, complexes of a Schiff base-18-
Page 2 of 11
1
2
3
4
5
6
crown-6 dinucleating ligand have recently been employed to mod-
ulate the electronic structure of CoII via cation complexation.18
Herein, we present a CoII-based PARACEST probe that can revers-
ibly bind Ca2+ and Na+ under physiological conditions. The ratio of
CEST signal intensities from the resulting Ca2+- and Na+-bound
probes enables, for the first time, the concentration-independent
quantitation of Ca2+ concentration by a MR-based method.
7
8
EXPERIMENTAL SECTION
Figure 1. Synthesis and molecular structure of LCo.
General Considerations. Unless otherwise specified, chemicals
and solvents were purchased from commercial vendors and used
without further purification. Deuterated solvents were purchased
from Cambridge Isotope Laboratories. When necessary for mois-
ture sensitive experiments, glassware was flame dried or stored in
an oven at 150 °C for at least 4 h, followed by cooling in a desic-
cator. Air- and water-free manipulations were carried out in a Vac-
uum Atmosphere Nexus II glovebox or using standard Schlenk
techniques under a dry dinitrogen atmosphere. Air-free experi-
ments involving the use of water were carried out in an MBraun
LABstar glovebox under a humid dinitrogen atmosphere. Acetoni-
trile (MeCN), diethyl ether (Et2O), and methanol (MeOH) were
dried using a commercial solvent purification system from Pure
Process Technology. MeCN was stored over 4 Å molecular sieves
prior to use. Water was obtained from a purification system from
EMD Millipore. Experimental details on the syntheses of organic
ligand precursors are provided in the Supporting Information.
Synthesis of 2,2'-(12,82-dihydroxy-9,12,15,18-tetraoxa-3,6-di-
aza-1,8(1,3)-dibenzenacyclooctadecaphane-3,6-diyl)diacetamide
(H2L, see Scheme S1). Under a dry dinitrogen atmosphere, 13,16,
19,22-tetraoxa-3,6-diazatricyclo[21.3.1.18,12]octacosa-1(27),8,10,
12(28),23,25-hexaene-27,28-diol (0.20 g, 0.48 mmol) and diiso-
propylethylamine (0.12 g, 0.96 mmol) were dissolved in MeCN (20
mL) to give a brown solution. The solution was heated at reflux and
2-bromoacetamide (0.13 g, 0.96 mmol) was added dropwise to the
boiling solution with vigorous stirring. After stirring at reflux for
12 h, basic alumina (2 g) was added and the mixture was evaporated
to dryness under reduced pressure. The resulting powder was dry-
loaded on a basic alumina column, which was packed using CH2Cl2
eluent. After loading, the column was first eluted with 2% (v/v)
MeOH/CH2Cl2 until no compound was detected by thin-layer chro-
matography (TLC), as visualized by I2 vapor, to remove an impu-
rity (Rf = 0.6 in 5% (v/v) MeOH/CH2Cl2). The column was then
eluted with 5% (v/v) MeOH/CH2Cl2 to obtain the desired product
(Rf = 0.3 in 5% (v/v) MeOH/CH2Cl2). The combined fractions were
evaporated to dryness under reduced pressure to give the product
as an off-white solid (25 mg, 10%). 1H NMR (MeOH-d4): 6.88 (t,
2H), 6.72 (m, 4H), 4.14 (m, 4H), 3.88 (m, 4H), 3.77 (s, 4H), 3.65
(s, 4H), 3.03 (s, 4H), 2.72 (s, 4H).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
into a pink solution of 1 in H2O over two weeks yielded single crys-
tals of LCo·0.50C3H6O·9.05H2O (1′) suitable for X-ray structural
analysis.
Synthesis of LCoNa(NO3)·1.7H2O (2). To a stirring pink solution
of 1 (30 mg, 0.046 mmol) in MeOH (5 mL), a solution of NaNO3
(4.3 mg, 0.051 mmol) in MeOH (0.5 mL) was added. The resulting
pink solution was stirred for 5 min at ambient temperature and then
filtered through diatomaceous earth. Slow diffusion of Et2O into
the pink solution over three days resulted in pink crystalline solid,
which was collected by filtration and dried in vacuo for 20 h to
afford 2 (34 mg, 95%). Anal. Calcd. for C26H37.4CoN5NaO12.7: C,
44.3; H, 5.35; N, 9.93. Found: C, 44.3; H, 5.16; N, 10.1. FT-IR
(ATR, cm–1): 3356 (s, broad); 2921 (m); 2873 (m); 1666 (s); 1591
(m); 1564 (m); 1476 (s); 1458 (s); 1302 (m); 1270 (m); 1231 (s);
1107 (s); 1086 (s); 987 (m); 956 (m); 930 (m); 896 (m); 840 (m);
741 (s); 409 (m). Slow diffusion of acetone into a pink solution of
2
in H2O over two weeks yielded single crystals of
[LCoNa(H2O)](NO3)0.5(OH)0.5·5.8H2O (2′) suitable for X-ray
structural analysis.
Synthesis of LCoCa(NO3)2·0.25Et2O·0.50H2O (3). To a stirring
pink solution of 1 (30 mg, 0.046 mmol) in MeOH (5 mL), a solution
of Ca(NO3)2·4H2O (12 mg, 0.051 mmol) in H2O (0.5 mL) was
added. The resulting pink solution was stirred for 5 min at ambient
temperature and then filtered through diatomaceous earth. Slow
diffusion of Et2O into the pink solution over four days resulted in
pink crystalline solid, which was collected by filtration and dried
in vacuo for 20 h to afford 3 (30 mg, 83%). Anal. Calcd. for
C27H37.5CaCoN6O14.75: C, 41.5; H, 4.84; N, 10.8. Found: C, 41.5;
H, 5.03; N, 10.9. FT-IR (ATR, cm–1): 3341 (s, broad); 2922 (m);
2874 (m); 1662 (s); 1600 (m); 1566 (m); 1475 (s); 1325 (s); 1303
(m); 1233 (s); 1105 (s); 1086 (s); 1071 (s); 1022 (m); 983 (m); 958
(m); 942 (m); 843 (m); 828 (m); 738 (s); 444 (w); 431 (m). Slow
diffusion of Et2O into a pink solution of 3 in MeOH over two weeks
yielded single crystals of [LCoCa(NO3)(MeOH)](NO3)·MeOH (3′)
suitable for X-ray structural analysis.
X-Ray Structure Determination. Single crystals of 1′, 2′, and
3′ were directly coated with Paratone-N oil and mounted on a Mi-
croMountsTM rod. The crystallographic data were collected at 100
K on a Bruker APEX II diffractometer equipped with a MoKα
sealed tube source. Raw data were integrated and corrected for Lo-
rentz and polarization effects using Bruker APEX2 v. 2009.1.19
The program SADABS was used to apply absorption correction.20
Space group assignments were determined by examining system-
atic absences, E-statistics, and successive refinement of the struc-
ture. Structures were solved by SHELXT21a,b using direct methods
and refined by SHELXL21a,b within the OLEX2 interface.21c All hy-
drogen atoms were placed at calculated positions using suitable rid-
ing models and refined using isotropic displacement parameters de-
rived from their parent atoms. Thermal parameters for all non-hy-
drogen atoms were refined anisotropically. The solvent mask pro-
cedure as implemented in OLEX2 was applied to the structures of
1′ and 2′ to account for severely disordered solvent molecules that
could not be properly modeled. Void volumes of 9484.8 and
11161.8 Å3 with a total of 4060.8 and 3043.0 electrons, respec-
tively, were found per unit cell in the crystal structures of 1′ and 2′,
Synthesis of LCo·3.1H2O (1, see Figure 1). Under a dinitrogen
atmosphere, H2L (25 mg, 0.047 mmol) was dissolved in MeOH (5
mL). To this colorless solution, a solution of Co(CH3COO)2·4H2O
(12 mg, 0.048 mmol) in MeOH (5 mL) was added. The resulting
light magenta solution was heated at reflux under a dinitrogen at-
mosphere for 12 h. The solution was evaporated to dryness and the
residue was dissolved in MeOH (2 mL). The pink solution was
added to Et2O (15 mL) with vigorous stirring, to induce the for-
mation of a pink precipitate. The pink solid was collected by filtra-
tion, washed with Et2O (5 mL), and dried in vacuo for 20 h to give
1 (25 mg, 82%). Anal. Calcd. for C26H40.2CoN4O11.1: C, 48.4; H,
6.28; N, 8.68. Found: C, 48.2; H, 5.62; N, 9.14. UV-visible absorp-
tion spectrum (H2O, 25 °C): 517 nm (e = 59.5 M−1 cm−1), 526 nm
(e = 57.7 M−1 cm−1). FT-IR (ATR, cm–1): 3341 (s, broad); 2916 (m);
2872 (m); 1662 (s); 1589 (s); 1562 (s); 1456 (s); 1304 (m); 1229
(s); 1110 (s); 1070 (s); 987 (m); 955 (m); 932 (m); 894 (m); 840
(m); 737 (s); 587 (w); 439 (w); 407 (w). Solution magnetic moment
(D2O, 37 °C): cMT = 2.4(3) cm3 K mol−1. Slow diffusion of acetone
2
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
the buffer solutions was adjusted to the desired values using aque-
respectively. These were ascribed to 8.5 and 6.3 H2O molecules per
1
2
3
4
5
6
7
8
LCo unit in the structures of 1′ and 2′, respectively. In the structure
of 2′, the occupancies of Na+ and NO3– ions within each asymmet-
ric unit were found to be 1.0 and 0.5, respectively. As such, 0.5
OH– ion per LCo unit is likely present for charge balancing. Crys-
tallographic data for 1′, 2′, and 3′ at 100 K and the details of data
collection are listed in Tables S1–S3.
ous HCl and NMe4OH solutions to avoid introduction of inorganic
cations. Chemical shift values (δ) are reported in ppm and refer-
enced to residual signals from the deuterated solvents (7.26 ppm
for CDCl3, 3.31 ppm for MeOH-d4, and 2.50 ppm for DMSO-d6).
For measurements of complexes in D2O or H2O, the chemical shift
of the solvent signal was set to 0 ppm to simplify comparison be-
tween 1H NMR and CEST spectra (see Figures S1 and S2).
Solid-State Magnetic Measurements. Magnetic measurements
of 1, 2, and 3 were performed on polycrystalline samples dispensed
in icosane. Samples were loaded in quartz tubes under a dinitrogen
atmosphere, attached to a sealable hose adapter, and flame sealed
under vacuum on a Schlenk manifold. All data were collected using
a Quantum Design MPMS-XL SQUID magnetometer. The reduced
magnetization data were collected between 1.8 and 10 K at applied
dc fields ranging from 0 to +7 T. The program PHI22 was employed
to simulate the reduced magnetization data using the following spin
Hamiltonian:
CEST Experiments. All CEST experiments were performed at
37 °C on an Agilent DD2 500 MHz (11.7 T) or a Varian Inova 500
MHz (11.7 T) systems. Samples for measurements contained 2.5–
11 mM of 1 in 50 mM HEPES buffer solutions at pH 7.4, in the
absence and presence of inorganic cations. All samples were pre-
pared and stored under dinitrogen atmosphere to ensure no degra-
dation due to oxidation by air. Z-spectra (CEST spectra) were ob-
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
tained according to the following protocol. H NMR spectra were
acquired from –100 to 100 ppm frequency offset (chemical shift
with respect to the bulk H2O signal) with a step increase of 1 ppm
using a presaturation pulse applied for 3 s at a power level (B1) of
21–22 µT. The B1 values were calculated based on the calibrated
90° pulse on a linear amplifier. D2O was placed in an inner capillary
within each NMR sample tube to lock the sample. The obtained 1H
NMR spectra were plotted as normalized integrations of the H2O
signal against frequency offset to produce a Z-spectrum. Direct sat-
uration of the H2O signal was set to 0 ppm. CEST peak intensities
from 20 to 40 ppm were fitted using a linear model to construct
baselines, based on which the relevant CEST intensities were cor-
rected. Exchange rate constants were calculated based on a reported
Ĥ = DŜz2 + gµBS·H (1)
In this Hamiltonian, D is the axial zero-field splitting parameter, Ŝz
is the z component of the spin angular momentum operator, g is the
electron spin g-factor, μ is the Bohr magneton, S is the spin angular
B
momentum, and H is the applied magnetic field. Isotropic g values
of 2.30(1), 2.28(1), and 2.33(1) were used for fitting the data for 1,
2, and 3, respectively, to extract the D values (see Table 1 and Fig-
ure 2).
Solution Magnetic Measurements. The solution magnetic mo-
ment of 1 was determined at 37 °C using the Evans method23 by
collecting 1H NMR spectra on an Agilent DD2 500 MHz (11.7 T)
system. Samples contained 5 mM of 1 in a mixture of 0.5% (w/w)
of dimethyl sulfoxide (DMSO) in D2O and were prepared under
dinitrogen atmosphere to ensure no degradation due to oxidation by
air. A capillary containing the same solvent mixture but without 1
was inserted into each NMR sample tube as a reference. Diamag-
netic corrections were carried out based on the empirical formula
of 1 using Pascal’s constants.24 The paramagnetic molar suscepti-
bility cMpara (cm3 mol−1) was calculated using the following equa-
1
method.25 In particular, H NMR spectra were acquired at various
presaturation power levels ranging from 7.4 to 21 µT applied for 6
s at 37 °C.
Determination of Dissociation Constants (Kd) by 1H NMR Ti-
1
tration Experiments. H NMR titration experiments were per-
formed following modified literature procedures.26 In order to ob-
tain accurate chemical shift and peak integration values, samples
were prepared in D2O under dinitrogen atmosphere. Reported val-
ues of Kd for each cation-bound complex are averages from two
independent experiments. For [LCoM]+/2+ (M+/2+ = Na+, Mg2+, K+),
tion:23
para
dia
1
cM
= (3DnMw)/(4πn0 m) − cM
(2)
values of Kd were estimated by monitoring changes in H NMR
chemical shift for two different resonances (see Table S4) and av-
erages from these data sets are reported in Table 2. To estimate Kd
for the interaction between LCo and Na+ and Mg2+ ions, samples
containing 2.5 or 1.0 mM of 1 were mixed with various amounts of
NaNO3 or Mg(NO3)2 to give final concentrations of Na+ and Mg2+
ranging from 0 to 18.7 mM and 0 to 56.7 mM, respectively. The
interconversion rate between {LCo + M+/2+} and [LCoM]+/2+ (M+/2+
= Na+, Mg2+) was fast compared to the 1H NMR acquisition time
scale (ca. 10–3 s), as evidenced by the presence of a single set of
NMR resonances for the whole series of spectra for both ions. The
changes in chemical shift (Dd) for the peaks at ca. 202 and 123 ppm
were monitored for each added ion (see Figures 3, left, and S3–S5)
and fitted using the program Dynafit27 to extract values of Kd (see
Script S1, and Tables 2 and S4). Representative fits of these data
are shown in Figures S6–S9.
Because Kd for [LCoK]+ was expected to be lower than that for
[LCoNa]+, 150 mM of Na+ was introduced to compete with K+ for
bindingto LCo, so an equilibrium could be established at a concen-
tration of 1 high enough to be observed by 1H NMR. Samples con-
taining 2.5 mM of 1 and 150 mM of NaNO3 in D2O were mixed
with various amounts of KNO3 to give final concentrations of K+
ranging from 0 to 18.6 mM. The interconversion rate between
{[LCoNa]+ + K+} and {[LCoK]+ + Na+} was fast compared to the
1H NMR acquisition time scale (ca. 10–3 s), as evidenced by the
presence of a single set of NMRresonances for the series of spectra.
The changes in chemical shift (Dd) for the peaks at ca. 212 and 133
ppm (see Figures S10 and S11) were fitted using Dynafit,27 using a
In this equation, Dn is the frequency difference (Hz) between the
DMSO resonance in the sample and reference solutions, Mw is the
molecular mass of the paramagnetic compound (g mol−1), n0 is the
operating frequency of the NMR spectrometer (Hz), m is the con-
centration of the paramagnetic compound (g cm−3), and cMdia is the
diamagnetic contribution to the molar susceptibility (cm3 mol−1).
The reported value of cMT is an average from three independent
measurements.
1H NMR Spectroscopy. 1H NMR spectra of H2L and ligand pre-
cursors were collected at 25 °C on an automated Agilent DD MR
400 MHz (9.4 T), an Agilent DD2 500 MHz (11.7 T), or on a Var-
ian Inova 500 MHz (11.7 T) spectrometers. All 1H NMR spectra of
cobalt complexes were recorded at 37 °C. Data for dissociation
constant (Kd) measurements in D2O were collected on a Bruker
Avance III HD Nanobay 400 MHz (9.4 T) system (for Na+ and K+)
or on a Bruker Neo 600 MHz (14.1 T) system equipped with a QCI-
F cryoprobe (for Mg2+ and Ca2+), and data for Kd measurements in
50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
(HEPES) buffer solutions at pH 7.3–7.5 were collected on an Ag-
ilent DD2 500 MHz (11.7 T) system. 1H NMR spectra of samples
of 1 in D2O and in 50 mM HEPES buffer solutions at pH 7.4 were
collected on a Bruker Avance III HD Nanobay 400 MHz (9.4 T)
system and on an Agilent DD2 500 MHz (11.7 T) system, respec-
tively. For the HEPES buffer solution samples, D2O was placed in
an inner capillary within each NMR sample tube to lock the sample.
Samples were prepared and stored under dinitrogen atmosphere to
ensure no degradation due to oxidation by air. Note that the pH of
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
competition model (see Script S2), to afford values of Kd (see Fig- CH Instruments 760 c potentiostat under a humid dinitrogen atmos-
1
2
3
4
5
6
7
8
ures S12 and S13, and Tables 2 and S4).
phere inside an MBraun LABstar glovebox at ambient temperature.
The cell consisted of a glassy carbon electrode as a working elec-
trode, a platinum wire as a counter electrode, and a saturated calo-
mel electrode (SCE) as a reference electrode. The analyte solutions
were prepared using 50 mM of HEPES buffered at pH 7.4 with ei-
ther 100 mM of NMe4Cl or a mixture of inorganic cations at their
physiological concentrations (150 mM of NaCl, 4 mM of KNO3, 2
mM of Ca(NO3)2, and 0.2 mM of Mg(NO3)2) as an electrolyte. The
voltammograms were converted and referenced to the normal hy-
drogen electrode (NHE), using a literature conversion factor.28
Other Physical Measurements. Infrared data were collected on
a Bruker Alpha FTIR spectrometer equipped with an attenuated to-
tal reflectance accessory. Solution UV-visible-NIR spectra were
obtained using an Agilent Cary 5000 spectrophotometer. Elemental
analyses of 1, 2, and 3 were performed by Midwest Microlab (In-
dianapolis, IN).
To determine Kd for [LCoCa]2+, K+ was used as a competing ion.
Samples containing 1.0 mM of 1 and 30 mM of KNO3 were mixed
with various amounts of Ca(NO3)2 to give final concentrations of
Ca2+ ranging from 0 to 3.65 mM. The increase in the intensity of
the peak at 245 ppm with increasing concentration of Ca2+ and the
concomitant decrease in signal intensity for the peak at 207 ppm
indicated a slow interconversion rate between {[LCoK]+ + Ca2+}
and {[LCoCa]2+ + K+} compared to the 1H NMR acquisition time
scale (ca. 10–3 s). The integration values for the peaks at 207 and
245 ppm were normalized to represent the relative percentages of
[LCoK]+ and [LCoCa]2+ in the samples (see Figure S14). The value
of Kd for [LCoCa]2+ was estimated using the following equation
(see Table 2):
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kd (LCoCa) = Kd (LCoK) × [Ca2+]/[K+] (3)
In this equation, Kd
is the average Kd value obtained for
(LCoK)
[LCoK]+, and [Ca2+] and [K+] are the concentrations of free Ca2+
and K+ ions when [LCoK]+ : [LCoCa]2+ = 1 : 1.
RESULTS AND DISCUSSION
Syntheses and Structures. The design of H2L was inspired by
previous reports of transition metal Schiff base complexes featur-
ing an appended crown ether pocket.16a,18,29 An SN2 reaction be-
tween 2-bromoacetamide and a reduced salen precursor afforded
H2L (see Scheme S1). The crown ether group is responsible for cat-
ion recognition, whereas the hexadentate chelating fragment fea-
tures exchangeable carboxamide protons and accommodates an an-
isotropic metal center suitable for PARACEST. CoII is an ideal
metal ion for these purposes owing to high magnetic anisotropy and
fast electronic relaxation time.17 The crown ether moiety was ex-
pected to be well suited for selective and reversible binding of Ca2+
under physiological conditions as it mimics the organic molecule
18-crown-6, which is well known to bind cations with different af-
finities based on their size and charge.30 Furthermore, structural
changes in the crown ether pocket caused by cation binding were
To investigate the effects of pH on cation binding to LCo and
compare to the data obtained in D2O, 1H NMR titration experiments
were repeated for Na+ in deoxygenated 50 mM HEPES buffer so-
lutions at pH 7.3–7.5. Samples containing 5.0 mM of 1 and 50 mM
of HEPES buffered at pH 7.3, 7.4, and 7.5, respectively, were
mixed with various amounts of NaNO3 to give final concentrations
of Na+ ranging from 0 to 18.2 mM. The change in chemical shift
(Dd) for the peak at ca. 123 ppm was monitored and fitted using a
non-competition model in Dynafit27 (see Script S1) to obtain values
of Kd as described above. These data are summarized in Figures
S15–S17 and Table S5.
Electrochemical Measurements. Cyclic voltammetry measure-
ments were carried out in a standard one-compartment cell using
Figure 2. Upper: Crystal structures of LCo (left), [LCoNa(H2O)]+ (middle), and [LCoCa(NO3)(MeOH)]+ (right), as observed in 1′, 2′, and 3′, respectively.
Magenta, cyan, lilac, red, blue, and gray spheres represent Co, Ca, Na, O, N, and C atoms, respectively; H atoms are omitted for clarity. Lower: Low-
temperature magnetization data for 1 (left), 2 (middle), and 3 (right), collected at selected dc fields (see legends). Colored circles and black solid lines represent
experimental data and corresponding fits, respectively.
4
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
ionic radii,31 the less distorted structure in 3′ than 2′ likely stems
Table 1. Selected mean interatomic distances and octahedral distortion pa-
rameter (S)31 for 1′–3′, and axial zero-field splitting parameter (D) for 1–3.
1
from greater electrostatic attraction between Ca2+ and the O atoms.
2
3
4
5
6
7
8
9
1′ / 1
2′ / 2
3′ / 3
The conformational differences between the crown ether units in
1′, 2′, and 3′ caused significant structural differences at the CoII
center in the three compounds. This effect can be quantified
through the octahedral distortion parameter (S), which is defined as
the sum of the absolute deviations from 90° for all 12 cis L–Co–L
angles.32 Across the series, 1′ features the largest distortion from an
octahedral geometry at CoII with S = 97.8(5)°, followed by 2′ with
S = 91.6(4)°, and 3′ with S = 66.3(2)° (see Table 1).
Co−Oamide (Å)
2.158(4)
2.00(2)
2.22(3)
97.8(5)
−18.7(3)
2.15(3)
2.009(8)
2.18(3)
91.6(4)
−20.8(2)
2.12(3)
Co−Ophenoxo (Å)
1.978(1)
2.141(5)
66.3(2)
Co−N (Å)
S (°)
D (cm−1)
−40.0(1)
Solid-State Magnetic Properties. Given the significant differ-
ences in coordination geometry at CoII across the three compounds,
one would expect associated changes in magnetic anisotropy. To
probe the influence of coordination geometry on magnetic anisot-
ropy in these compounds, low-temperature magnetization data
were collected for polycrystalline samples of 1, 2, and 3 at selected
dc fields (see Figure 2, lower). The non-superimposability of the
resulting isofield curves for all compounds, along with their satu-
ration magnetization values below M = 3 µB expected for an iso-
tropic S = 3/2 CoII center, indicates the presence of significant zero-
field splitting, which is a measure of magnetic anisotropy. This ef-
fect was quantified by fitting the data using equation 1 (see Exper-
imental Section),22 giving axial zero-field splitting parameters of D
= −18.7(3), −20.8(2), and −40.0(1) cm−1 for 1, 2, and 3, respec-
tively. Here, the magnitude of D increases with decreasing distor-
tion from octahedral geometry at CoII, which is in line with a pro-
gression toward orbital degeneracy in moving from 1 to 2 to 3.
Solution 1H NMR Properties. To probe how changes in mag-
netic anisotropy of CoII affect the NMR hyperfine shifts of ligand
protons in 1, 2, and 3, 1H NMR spectra were collected at 37 °C for
solutions containing 5 mM of 1 and 50 mM of HEPES buffered at
pH 7.4, in the absence and presence of 15 mM of NaNO3 or
Ca(NO3)2. Note that excess amounts of Na+ and Ca2+ were used to
ensure complete cation binding, and no further spectral changes
were observed beyond this concentration. Spectra for all three so-
lutions display sharp peaks spanning from −23 to 245 ppm vs H2O,
consistent with high-spin CoII in all compounds (see Figure S1).
Carboxamide resonances were observed at 77, 69, and 80 ppm for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
envisioned to influence the coordination geometry of the hexaden-
tate chelate, and thus affect the magnetic anisotropy of the para-
magnetic CoII center. These changes in magnetic anisotropy of CoII
were anticipated to impact the hyperfine shifts of carboxamide pro-
tons from the pendent donors, providing CEST peaks with cation-
dependent frequencies.
Reaction of H2L with Co(CH3COO)2·4H2O16a,18,29 afforded the
pink compound LCo·3.1H2O (1) in 82% yield (see Figure 1). Sub-
sequent addition of stoichiometric amounts of NaNO3 or
Ca(NO3)2·4H2O yielded the compounds LCoNa(NO3)·1.7H2O (2)
or LCoCa(NO3)2·0.25Et2O·0.50H2O (3), respectively. Slow diffu-
sion of acetone into aqueous solutions of 1 or 2, or Et2O into a so-
lution of 3 in MeOH, gave pink block-shaped crystals of
LCo·0.50C3H6O·9.05H2O (1′), [LCoNa(H2O)](NO3)0.5(OH)0.5·5.8
H2O (2′), and [LCoCa(NO3)(MeOH)](NO3)·MeOH (3′), respec-
tively. Single-crystal X-ray diffraction analyses of 1′–3′ (see Ta-
bles S1–S3) revealed that the CoII ion resides in a distorted octahe-
dral environment in all three structures, with the N2O2 pocket of
L2− comprising the equatorial plane and the O atoms from the pen-
dent carboxamide groups coordinating the axial sites (see Figure 2,
upper). In 2′, a Na+ ion is ligated by four of the six O atoms from
the crown ether unit of L2− and a H2O molecule to give an irregular
five-coordinate complex. In stark contrast, 3′ features a nine-coor-
dinate Ca2+ ion that induces only a minimal distortion to the crown
ether, where all six Ca–Ocrown distances are shorter than 2.74 Å. The
remaining three coordination sites of Ca2+ are occupied by a MeOH
–
molecule and an h2-NO3 ion. Because Ca2+ and Na+ have similar
1
Figure 3. Left: Change in H NMR chemical shift of a selected resonance for a 2.5 mM solution of 1 in D2O upon incremental addition of NaNO3. Right:
Changes in 1H NMR signal intensities of selected resonances for a D2O solution containing 1.0 mM of 1 and 30 mM of KNO3 upon incremental addition of
Ca(NO3)2. The resonances at 207 and 245 ppm correspond to [LCoK]+ and [LCoCa]2+, respectively. Data were collected at 37 °C at 9.4 and 14.1 T for Na+
and Ca2+, respectively. Numbers next to spectra denote the concentrations of respective added cations (mM).
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
Table 2. Summary of dissociation constants (Kd) for cation-bound com-
plexes of 1 in D2O at 37 °C.
1
2
[LCoNa]+
[LCoMg]2+
23(2)a
[LCoK]+
0.3(2)a
[LCoCa]2+
0. 01(1)b
3
4
5
6
Kd (mM)
4.8(3)a
a Average value from monitoring 1H NMR chemical shift changes upon cat-
ion addition for two different resonances. b Estimated by a method de-
scribed in the Experimental Section.
7
8
9
1, [LCoNa]+, and [LCoCa]2+, respectively, as evidenced by their
disappearance in the spectra recorded in D2O (see Figure S2). The
observation of a single carboxamide peak for all compounds sug-
gests chemical equivalence of the two carboxamide groups in each
molecule. Importantly, the difference in chemical shift of 11 ppm
between the Ca2+- and Na+-bound compounds is more than two or-
ders of magnitude greater than that of a diamagnetic analogue,18
highlighting the high sensitivity of 1H NMR hyperfine shift toward
structural and magnetic differences at CoII. Note that the carbox-
amide peak is not the most shifted resonance for any of the three
compounds. The most shifted resonances for 1, [LCoNa]+, and
[LCoCa]2+ are located at 203, 210, and 245 ppm, respectively. The
increase in maximum hyperfine shift in moving from 1 to [LCoNa]+
to [LCoCa]2+ is in good agreement with the increase in magnetic
anisotropy across the series, as evident from solid-state magnetic
measurements. The 35 ppm difference in maximum hyperfine shift
between [LCoCa]2+ and [LCoNa]+ represents even higher cation
sensitivity, suggesting that there is a large room for improving the
sensitivity of cation-sensing MR probes.
Assessment of Cation Binding Affinities by 1H NMR. While
UV-visible absorption spectroscopy is commonly used to deter-
mine dissociation constants, the UV-visible spectra for aqueous so-
lutions of 1, [LCoNa]+, and [LCoCa]2+ reveal no significant differ-
ences (see Figure S18). However, because notable changes were
observed between the 1H NMR spectra for these compounds, and
1H NMR has been employed in studying cation binding for other
crown ether-based systems,26 we decided to assess the binding af-
finities of 1 toward Na+, Mg2+, K+, and Ca2+ through 1H NMR titra-
tion experiments at 37 °C. Addition of Na+ or Mg2+ to solutions of
1 in D2O resulted in downfield shifting of 1H NMR resonances. The
resonances shifted non-linearly with increasing concentrations of
Na+ or Mg2+, suggesting the presence of equilibrium (see Figures
3, left, and S3–S5). The changes in chemical shift for the peaks at
ca. 202 and 123 ppm were most pronounced and could be mod-
eled27 to provide average dissociation constants of Kd = 4.8(3) and
23(2) mM for [LCoNa]+ and [LCoMg]2+, respectively (see Figures
3, left, and S3–S9, and Table S4). Owing to the structural similarity
between 18-crown-6 and the crown ether moiety of H2L, 1 was ex-
pected to display higher affinity toward K+ than toward the smaller
ions Na+ and Mg2+.18,29 As such, Na+ was introduced to compete
with K+ for binding to LCo, so an equilibrium could be established
at a concentration of 1 high enough to be observed by 1H NMR. A
similar non-linear shift of 1H resonances was observed upon incre-
mental addition of KNO3 to a D2O solution of 1 containing 150 mM
of NaNO3 as observed in the Na+ and Mg2+ titration experiments,
albeit less pronounced (see Figures S10 and S11). A fit of the data
gave an average value of Kd = 0.3(2) mM for [LCoK]+ (see Figures
S12 and S13, and Table S4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. CEST spectra collected at 11.7 T and 37 °C using a 3 s presatu-
ration pulse and B1 = 21 μT for 5 mM aqueous solutions of 1 containing 50
mM of HEPES buffered at pH 7.4 (black), and with 15 mM of Na+ (red) or
Ca2+ (blue). Inset: Expanded view of relevant CEST peaks.
Figure S14). By using the previously determined Kd value for
[LCoK]+ and [Ca2+] at 50% [LCoCa]2+ formation, a value of Kd =
0.01(1) mM was estimated for [LCoCa]2+ (see equation 3). Average
values of Kd for the different cation-bound complexes of 1 are sum-
marized in Table 2. Cation affinity for 1 follows the order Ca2+
>
K+ > Na+ > Mg2+. The tightest binding of 1 to Ca2+ is likely a result
of an optimal ionic radius in conjunction with a high positive
charge. Most importantly, these Kd values suggest that 1 will near
exclusively bind Ca2+ and Na+ over Mg2+ and K+ under physiolog-
ical conditions owing to low affinities and/or low concentrations of
the latter two ions. Furthermore, similar values of Kd = 1.8(9)–
3.7(9) mM were obtained for [LCoNa]+ in HEPES solutions buff-
ered at pH 7.3–7.5 as observed in D2O (see Figures S15–S17 and
Table S5), indicating that ion binding to 1 is minimally affected by
pH in the physiological range.
CEST Properties. To investigate the potential of 1 as a cation-
responsive PARACEST probe, CEST spectra were collected at 37
°C for solutions containing 5 mM of 1 and 50 mM of HEPES buff-
ered at pH 7.4, in the absence and presence of 15 mM of Na+ or
Ca2+. For solutions of 1, [LCoNa]+, and [LCoCa]2+, CEST peaks
were observed at 77, 69, and 80 ppm, respectively, with 4.8, 3.9,
and 8.5% H2O signal reduction, respectively (see Figure 4). These
CEST peak frequencies are consistent with the assignment of car-
1
boxamide resonances from the H NMR spectra. Despite the low
signal intensities, the frequency difference between the three CEST
peaks highlights the effectiveness of LCo to distinguish between
Na+ and Ca2+ in solution. A second CEST peak was observed at 11
ppm in the spectrum for [LCoCa]2+, likely stemming from a coor-
dinating H2O molecule. Exchange rates for the carboxamide pro-
tons were estimated by the Omega plot method at 37 °C and pH
7.4.25 Rate constants of kex = 4.0(5) × 102, 3.0(6) × 102, and 2.4(2)
× 102 s–1 were obtained for 1, [LCoNa]+, and [LCoCa]2+, respec-
tively, (see Figure S19). These values are in good agreement with
those reported for other carboxamide-based PARACEST
agents.9,10c,11a,c,h–k The stronger CEST effect observed for
[LCoCa]2+ compared to that for [LCoNa]+, despite similar values
of kex for the two complexes, could be due to the presence of a third
pool of labile protons from the coordinating H2O molecule in
HEPES buffer solutions of [LCoCa]2+. This hypothesis is supported
by the observation of an additional peak at 11 ppm in the CEST
spectrum for [LCoCa]2+.
For the NMR titrations of LCo with Na+, Mg2+, and K+, only one
set of 1H resonances was observed in each case, indicating fast cat-
ion exchange rates compared to the 1H NMR acquisition time scale
(ca. 10−3 s). In contrast, when Ca2+ was added to a D2O solution of
1 containing K+ as a competing ion, two sets of 1H resonances were
observed (see Figure 3, right). This observation suggests a slow
cation exchange between [LCoK]+ and [LCoCa]2+ in aqueous solu-
tions. Integrations of the peaks at 207 and 245 ppm, corresponding
to [LCoK]+ and [LCoCa]2+, respectively, could be employed to de-
rive the mole fraction of [LCoCa]2+ as a function of [Ca2+] (see
Quantitation of Ca2+ Concentration in Aqueous Solutions. To
evaluate the ability of 1 to enable a ratiometric quantitation of the
6
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
concentration of Ca2+ under physiological conditions, CEST spec-
tra were collected at 37 °C for a solution containing 2.0 mM of 1,
1
150 mM of NaCl, and 50 mM of HEPES buffered at pH 7.4, upon
incremental addition of Ca(NO3)2. In the absence of Ca2+, a single
CEST peak at 69 ppm was observed, indicating complete formation
of [LCoNa]+. However, upon addition of Ca2+, a new peak ap-
peared at 80 ppm, corresponding to [LCoCa]2+. The intensity of this
peak increased monotonically until reaching saturation at [Ca2+] =
3.13 mM. Due to partial overlap with the peak at 80 ppm, the CEST
peak intensity at 69 ppm first increased until [Ca2+] = 1.00 mM, but
then decreased substantially with increasing [Ca2+], reaching a
value of less than 1% at [Ca2+] = 3.13 mM (see Figure 5, upper).
The appearance of the CEST peak at 80 ppm demonstrates that 1
selectively binds Ca2+ over Na+ under physiological conditions.14
Importantly, the intensity of the [LCoCa]2+ CEST peak at 80 ppm
reached saturation at [Ca2+] = 3.13 mM rather than 2.00 mM, indi-
cating that cation selectivity is modest enough to allow for an equi-
librium between [LCoNa]+ and [LCoCa]2+, thus enabling rati-
ometric quantitation of [Ca2+]. These observations are consistent
with the Kd values estimated for the two complexes.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To assess the influence of probe concentration on the CEST
properties of 1, variable-[Ca2+] CEST spectra were collected for
various concentrations of 1 in analogy to the 2.0 mM sample (see
Figures S20–S22). The intensities of the CEST peaks at 80 and 69
ppm changed considerably as the concentration of 1 was varied,
demonstrating the shortcoming of detecting [Ca2+] solely based on
CEST peak intensities. However, the ratio of peak intensities at 80
and 69 ppm (CEST80 ppm/CEST69 ppm) was only minimally affected
by the concentration of 1 (see Figure 5, lower). Note that the CEST
effect at 69 ppm cannot be accurately determined when [Ca2+] > 3
mM owing to the low intensity (see Figure 5, upper). As such,
CEST80 ppm/CEST69 ppm values for [Ca2+] > 3 mM may contain sig-
nificant error. The correlation between CEST80 ppm/CEST69 ppm and
[Ca2+] for [Ca2+] < 3 mM can be fitted using the following empirical
exponential model:
Figure 5. Upper: CEST spectra collected at 11.7 T and 37 °C using a 3 s
presaturation pulse and B1 = 21 μT for 50 mM HEPES buffer solutions con-
taining 2.0 mM of 1 and 150 mM of NaCl at pH 7.4 with increasing [Ca2+].
The legend denotes [Ca2+] (mM). Lower: Ratio of CEST peak intensities
from presaturation at 80 and 69 ppm vs [Ca2+]. Inset: Expanded view of the
relevant data. Circles and solid lines represent experimental data and fits,
respectively. The legend denotes [1] (mM).
CEST80 ppm/CEST69 ppm = exp([Ca2+] – x) (4)
Fits of the CEST data to equation 4 afforded x = 0.92(4), 0.90(6),
1.6(1), and 2.8(2) for 2.0, 2.8, 5.6, and 11 mM of 1, respectively.
Therefore, when only data for [Ca2+] < 3 mM are examined, the
equations for 2.0 and 2.8 mM of 1 are statistically indistinguishable
(see Figure 5, lower, inset). However, the equations for 5.6 and 11
mM of 1 are significantly different, suggesting that [LCoNa]+ and
[LCoCa]2+ are not in equilibrium at higher concentrations of 1 (≥
5.6 mM), likely due to strong Ca2+ binding. These results establish
the validity of using the CEST peak intensity ratio to quantitate
[Ca2+] independent of the concentration of 1 within a regime where
the concentration of 1 is sufficiently low (< 3 mM) to allow for an
equilibrium between [LCoNa]+ and [LCoCa]2+. Accordingly, to ex-
pand the range of concentration-independent quantitation of [Ca2+]
using PARACEST, the probe should exhibit weaker binding affin-
ity toward Ca2+ and stronger CEST effects.
To further test the feasibility of 1 to quantitate [Ca2+] in physio-
logical environments, variable-[Ca2+] CEST experiments were car-
ried out at 37 °C for a pH 7.4 buffer solution containing 2.8 mM of
1, 150 mM of Na+, 4 mM of K+, and 0.2 mM of Mg2+ to mimic their
physiological concentrations14 (see Figure S23). Prior to addition
of Ca2+, a single CEST peak at 69 ppm was observed with 0.8%
CEST effect, suggesting that LCo exclusively bound Na+. Upon in-
cremental addition of Ca2+, the intensity of this peak first increased
to 4% at [Ca2+] = 1.01 mM and then decreased monotonically with
increasing [Ca2+], whereas a CEST signal at 80 ppm corresponding
to [LCoCa]2+ appeared and reached a maximum intensity of 6.5%
at [Ca2+] = 3.40 mM. These spectral changes are analogous to those
observed in the absence of K+ and Mg2+. An equilibrium between
[LCoCa]2+ and [LCoNa]+ was again established, as the CEST peak
at 80 ppm reached a maximum intensity at [Ca2+] > 2.8 mM. How-
ever, both CEST peaks were noticeably broader compared to those
observed in the absence of K+ and Mg2+. Specifically, the two peaks
start to coalesce at [Ca2+] > 1 mM, as evidenced by the downfield
and upfield shifting of the CEST peaks at 69 and 80 ppm, respec-
tively. Together, these observations suggest an accelerated cation
exchange between [LCoCa]2+ and [LCoNa]+, likely stemming from
the presence of K+ andMg2+.
The plot of CEST80 ppm/CEST69 ppm vs [Ca2+], as depicted in Fig-
ure S24, reveals that CEST80 ppm/CEST69 ppm increases with increas-
ing [Ca2+], following a similar trend as the data shown in Figure 5,
lower. Fits of the data to equation 4 using [Ca2+] < 2.5 mM gave x
= 1.56(8) (see Figure S25), which is significantly higher than the
value of x = 0.90(6) obtained for the 2.8 mM sample of 1 in the
absence of K+ and Mg2+. This discrepancy can likely be attributed
to the increased cation exchange rate between [LCoCa]2+ and
[LCoNa]+ in the presence of K+ and Mg2+. Note that the empirical
exponential model provided in equation 4 does not fit the CEST80
ppm/CEST69 ppm vs [Ca2+] data for the physiological ion mixture very
well. However, an alternative exponential model given in equation
5 affords much better agreement, providing parameters of a =
0.38(6) and b = 0.73(7) (see Figure S26).
CEST80 ppm/CEST69 ppm = a × exp(b × [Ca2+]) (5)
Taken together, variable-[Ca2+] CEST experiments for solutions
of 1 with different cations confirm that 1 selectively and reversibly
binds Ca2+ and Na+ over the related cations K+ and Mg2+, verifying
the feasibility of ratiometric quantitation of [Ca2+] under physio-
logical conditions.14 While interference from binding other cations
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
is insignificant, the rate of cation exchange was found to influence
Corresponding Author
1
2
3
4
5
6
the CEST peak intensity ratio, highlighting the necessity of con-
structing a calibration curve under conditions that strongly resem-
ble the targeted environment. As such, the effects of cation ex-
change rates on the CEST properties of different cation-bound
probes should be strongly considered in the design of future cation-
responsive CEST probes.
dharris@northwestern.edu
ORCID
Kang Du: 0000-0001-9947-2320
Agnes E. Thorarinsdottir: 0000-0001-9378-4454
T. David Harris: 0000-0003-4144-900x
Stability Studies. Finally, we sought to investigate the stability
of 1 in aqueous solutions. Cyclic voltammetry experiments were
carried out for solutions of 1 with 50 mM of HEPES buffered at pH
7.4 in the absence and presence of a mixture of Na+, K+, Mg2+, and
Ca2+ ions at their physiological concentrations.14 In the absence of
the inorganic cations, 1 exhibits one pseudo-reversible redox pro-
cess at 440 mV vs NHE that is assigned to the CoII/III potential (see
Figure S27). Surprisingly, the pseudo-reversible CoII/III redox event
was shifted cathodically to 374 mV vs NHE in the presence of the
cation mixture (see Figure S28). The more reductive potential ob-
served in the presence of the inorganic cations could be attributed
to the different electrolytes in the sample solutions. Nevertheless,
both potentials are more reductive than that for the reduction of O2
to H2O in a neutral solution,28 suggesting that 1 is susceptible to
oxidation in air. To further study the stability of 1 under aerobic
conditions, a solution of 1 buffered at pH 7.4 was prepared under
dinitrogen atmosphere and exposed to air while a UV-visible ab-
sorption spectrum was recorded at regular intervals (see Figure
S29). Over a 24 h period, the intensities of the absorption bands
centered at 247 and 296 nm gradually decreased, likely correspond-
ing to the oxidation of Co2+ to Co3+ over the course of hours.
7
8
9
Notes
The authors declare no competing financial interest.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENTS
This research was funded by the Air Force Research Laboratory
under agreement no. FA8659-15-2-5518 and Northwestern Uni-
versity.
REFERENCES
(1) (a) Coleman, R. E. Metastatic Bone Disease: Clinical Features, Patho-
physiology and Treatment Strategies. Cancer Treat. Rev. 2001, 27, 165–
176. (b) Mundy, G. R. Metastasis to Bone: Causes, Consequences and Ther-
apeutic Opportunities. Nat. Rev. Cancer 2002, 2, 584–593. (c) Kyle, R. A.;
Rajkumar, S. V. Multiple Myeloma. N. Engl. J. Med. 2004, 351, 1860–
1873.
(2) (a) Fraser, W. D. Hyperparathyroidism. Lancet 2009, 374, 145–158. (b)
Blaine, J.; Chonchol, M.; Levi, M. Renal Control of Calcium, Phosphate,
and Magnesium Homeostasis. Clin. J. Am. Soc. Nephrol. 2015, 10, 1257–
1272.
(3) (a) Ralston, S. H. Paget’s Disease of Bone. N. Engl. J. Med. 2013, 368,
644–650. (b) Tuck, S. P.; Layfield, R.; Walker, J.; Mekkayil, B.; Francis,
R. Adult Paget’s Disease of Bone: A Review. Rheumatology 2017, 56,
2050–2059.
(4) Minisola, S.; Pepe, J.; Piemonte, S.; Cipriani, C. The Diagnosis and
Management of Hypercalcaemia. BMJ 2015, 350, h2723.
CONCLUSIONS AND OUTLOOK
The foregoing results demonstrate the feasibility of quantitating
Ca2+ concentration in a ratiometric manner through cation-depend-
ent CEST effects. Importantly, the LCo probe features both a se-
lective and a reversible binding of Ca2+ and Na+ in the presence of
similar cations. The CEST peaks for the Ca2+- and Na+-bound
probes are highly shifted and can be distinctively addressed, thanks
to the sensitivity of 1H NMR hyperfine shift and magnetic anisot-
ropy to changes in the coordination environment at the paramag-
netic CoII center. Moreover, variable-[Ca2+] CEST experiments
confirmed the ability of LCo to quantitate [Ca2+], independent of
the probe concentration when an equilibrium between [LCoCa2+]
and [LCoNa+] was achieved ([LCo] < 3 mM and [Ca2+] < 3 mM).
(5) Ziegler, R. Hypercalcemic Crisis. J. Am. Soc. Nephrol. 2001, 12, S3–
S9.
(6) (a) Lauffer, R. B. Paramagnetic Metal Complexes as Water Proton Re-
laxation Agents for NMR Imaging: Theory and Design. Chem. Rev. 1987,
87, 901–927. (b) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B.
Gadolinium(III) Chelates as MRI Contrast Agents: Structure, Dynamics,
and Applications. Chem. Rev. 1999, 99, 2293–2352. (c) Caravan, P. Strate-
gies for Increasing the Sensitivity of Gadolinium Based MRI Contrast
Agents. Chem. Soc. Rev. 2006, 35, 512–523.
(7) (a) Li, W.; Fraser, S. E.; Meade, T. J. A Calcium-Sensitive Magnetic
Resonance Imaging Contrast Agent. J. Am. Chem. Soc. 1999, 121, 1413–
1414. (b) Li, W.; Parigi, G.; Fragai, M.; Luchinat, C.; Meade, T. J. Mecha-
nistic Studies of a Calcium-Dependent MRI Contrast Agent. Inorg. Chem.
2002, 41, 4018–4024. (c) Atanasijevic, T.; Shusteff, M.; Fam, P.; Jasanoff,
A. Calcium-Sensitive MRI Contrast Agents Based on Superparamagnetic
Iron Oxide Nanoparticles and Calmodulin. Proc. Natl. Acad. Sci. U. S. A.
2006, 103, 14707–14712. (d) Mishra, A.; Fousková, P.; Angelovski, G.;
Balogh, E.; Mishra, A. K.; Logothetis, N. K.; Tóth, É. Facile Synthesis and
Relaxation Properties of Novel Bispolyazamacrocyclic Gd3+ Complexes:
An Attempt towards Calcium-Sensitive MRI Contrast Agents. Inorg.
Chem. 2008, 47, 1370–1381. (e) Dhingra, K.; Maier, M. E.; Beyerlein, M.;
Angelovski, G.; Logothetis, N. K. Synthesis and Characterization of a
Smart Contrast Agent Sensitive to Calcium. Chem. Commun. 2008, 0,
3444–3446. (f) Angelovski, G.; Fousková, P.; Mamedov, I.; Canals, S.;
Tóth, É.; Logothetis, N. K. Smart Magnetic Resonance Imaging Agents that
Sense Extracellular Calcium Fluctuations. ChemBioChem 2008, 9, 1729–
1734. (g) Que, E. L.; Chang, C. J. Responsive Magnetic Resonance Imaging
Contrast Agents as Chemical Sensors for Metals in Biology and Medicine.
Chem. Soc. Rev. 2010, 39, 51–60. (h) Mamedov, I.; Canals, S.; Henig, J.;
Beyerlein, M.; Murayama, Y.; Mayer, H. A.; Logothetis, N. K.; Angelovski,
G. In Vivo Characterization of a Smart MRI Agent That Displays an Inverse
Response to Calcium Concentration. ACS Chem. Neurosci. 2010, 1, 819–
828. (i) Heffern, M. C.; Matosziuk, L. M.; Meade, T. J. Lanthanide Probes
for Bioresponsive Imaging. Chem. Rev. 2014, 114, 4496–4539. (j) Kadjane,
P.; Platas-Iglesias, C.; Boehm-Sturm, P.; Truffault, V.; Hagberg, G. E.;
Hoehn, M.; Logothetis, N. K.; Angelovski, G. Dual-Frequency Calcium-
Responsive MRI Agents. Chem. Eur. J. 2014, 20, 7351–7362. (k) Mac-
Renaris, K. W.; Ma, Z.; Krueger, R. L.; Carney, C. E.; Meade, T. J. Cell-
The current proof-of-principle study represents a first step to-
ward practical quantitation of [Ca2+], and potentially other physio-
logically relevant cations, in a concentration-independent manner.
While the weak CEST effects and air-sensitivity may preclude the
practical use of the current LCo probe, we have established that the
combination of magnetic anisotropy modulation and cation recog-
nition by a crown ether moiety is a promising cation-sensing strat-
egy. Future efforts will be directed toward tuning the transition
metal and pendent CEST-active groups to optimize sensitivity, cat-
ion binding affinities, CEST peak intensities, and probe stability,
as well as modifying the crown ether unit toward sensing cations of
different sizes.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website.
Additional experimental details and characterization
data (PDF)
Crystallographic data for 1′, 2′, and 3′ (CIF)
AUTHOR INFORMATION
8
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
Permeable Esterase-Activated Ca(II)-Sensitive MRI Contrast Agent. Bio-
conjugate Chem. 2016, 27, 465–473.
Menon, R. S.; Bartha, R. A Sensitive PARACEST Contrast Agent for Tem-
perature MRI: Eu3+-DOTAM-Glycine(Gly)-Phenylalanine(Phe). Magn.
Reson. Med. 2008, 59, 374–381. (c) Delli Castelli, D.; Terreno, E.; Aime,
S. Yb(III)-HPDO3A: A Dual pH- and Temperature-Responsive CEST
Agent. Angew. Chem., Int. Ed. 2011, 50, 1798–1800. (d) Jeon, I.-R.; Park,
J. G.; Haney, C. R.; Harris, T. D. Spin Crossover Iron(II) Complexes as
PARACEST MRI Thermometers. Chem. Sci. 2014, 5, 2461–2465.
(13) (a) Trokowski, R.; Ren, J.; Kálmán, F. K.; Sherry, A. D. Selective
Sensing of Zinc Ions with a PARACEST Contrast Agent. Angew. Chem.,
Int. Ed. 2005, 44, 6920–6923. (b) Srivastava, K.; Ferrauto, G.; Harris, S.
M.; Longo, D. L.; Botta, M.; Aime, S.; Pierre, V. C. Complete On/Off Re-
sponsive ParaCEST MRI Contrast Agents for Copper and Zinc. Dalton
Trans. 2018, 47, 11346–11357.
(14) (a) Wan, Q. J.; Kubáň, P.; Tanyanyiwa, J.; Rainelli, A.; Hauser, P. C.
Determination of Major Inorganic Ions in Blood Serum and Urine by Ca-
pillary Electrophoresis with Contactless Conductivity Detection. Anal.
Chim. Acta 2004, 525, 11–16. (b) Le, T.; Bhushan, V.; Sochat, M.; Chavda,
Y.; Zureick, A.; Kalani, M.; Kallianos, K. First Aid for the USMLE Step 1:
2018 a Student-to-Student Guide, 28th ed.; McGraw-Hill Education: New
York, 2017.
1
2
3
4
5
6
7
8
(8) Selected references: (a) Zhang, S.; Michaudet, L.; Burgess, S.; Sherry,
A. D. The Amide Protons of an Ytterbium(III) Dota Tetraamide Complex
Act as Efficient Antennae for Transfer of Magnetization to Bulk Water. An-
gew. Chem., Int. Ed. 2002, 41, 1919–1921. (b) Aime, S.; Delli Castelli, D.;
Terreno, E. Novel pH-Reporter MRI Contrast Agents. Angew. Chem., Int.
Ed. 2002, 41, 4334–4336. (c) Terreno, E.; Delli Castelli, D.; Cravotto, G.;
Milone, L.; Aime, S. Ln(III)-DOTAMGly Complexes: A Versatile Series
to Access the Determinants of the Efficacy of Paramagnetic Chemical Ex-
change Saturation Transfer Agents for Magnetic Resonance Imaging Ap-
plications. Invest. Radiol. 2004, 39, 235–243. (d) Woods, M.; Woessner, D.
E.; Zhao, P.; Pasha, A.; Yang, M.-Y.; Huang, C.-H.; Vasalitiy, O.; Morrow,
J. R.; Sherry, A. D. Europium(III) Macrocyclic Complexes with Alcohol
Pendant Groups as Chemical Exchange Saturation Transfer Agents. J. Am.
Chem. Soc. 2006, 128, 10155–10162. (e) Angelovski, G.; Chauvin, T.;
Pohmann, R.; Logothetis, N. K.; Tóth, É. Calcium-Responsive Paramag-
netic CEST Agents. Bioorg. Med. Chem. 2011, 19, 1097–1105.
(9) Selected references: (a) Dorazio, S. J.; Tsitovich, P. B.; Siters, K. E.;
Spernyak, J. A.; Morrow, J. R. Iron(II) PARACEST MRI Contrast Agents.
J. Am. Chem. Soc. 2011, 133, 14154–14156. (b) Tsitovich, P. B.; Morrow,
J. R. Macrocyclic Ligands for Fe(II) ParaCEST and Chemical Shift MRI
Contrast Agents. Inorg. Chim. Acta 2012, 393, 3–11. (c) Olatunde, A. O.;
Dorazio, S. J.; Spernyak, J. A.; Morrow, J. R. The NiCEST Approach:
Nickel(II) ParaCEST MRI Contrast Agents. J. Am. Chem. Soc. 2012, 134,
18503–18505. (d) Dorazio, S. J.; Morrow, J. R. Iron(II) Complexes Con-
taining Octadentate Tetraazamacrocycles as ParaCEST Magnetic Reso-
nance Imaging Contrast Agents. Inorg. Chem. 2012, 51, 7448–7450. (e)
Dorazio, S. J.; Olatunde, A. O.; Spernyak, J. A.; Morrow, J. R. CoCEST:
Cobalt(II) Amide-Appended ParaCEST MRI Contrast Agents. Chem. Com-
mun. 2013, 49, 10025–10027. (f) Du, K.; Harris, T. D. A CuII2 Paramagnetic
Chemical Exchange Saturation Transfer Contrast Agent Enabled by Mag-
netic Exchange Coupling. J. Am. Chem. Soc. 2016, 138, 7804–7807.
(10) (a) Tsitovich, P. B.; Spernyak, J. A.; Morrow, J. R. A Redox-Activated
MRI Contrast Agent That Switches Between Paramagnetic and Diamag-
netic States. Angew. Chem., Int. Ed. 2013, 52, 13997–14000. (b) Tsitovich,
P. B.; Burns, P. J.; McKay, A. M.; Morrow, J. R. Redox-Activated MRI
Contrast Agents Based on Lanthanide and Transition Metal Ions. J. Inorg.
Biochem. 2014, 133, 143–154. (c) Du, K.; Waters, E. A.; Harris, T. D. Ra-
tiometric Quantitation of Redox Status with a Molecular Fe2 Magnetic Res-
onance Probe. Chem. Sci. 2017, 8, 4424–4430.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(15) Bar-Shir, A.; Gilad, A. A.; Chan, K. W. Y.; Liu, G.; van Zijl, P. C. M.;
Bulte, J. W. M.; McMahon, M. T. Metal Ion Sensing Using Ion Chemical
Exchange Saturation Transfer 19F Magnetic Resonance Imaging. J. Am.
Chem. Soc. 2013, 135, 12164–12167.
(16) (a) van Staveren, C. J.; van Eerden, J.; van Veggel, F. C. J. M.;
Harkema, S.; Reinhoudt, D. N. Cocomplexation of Neutral Guests and Elec-
trophilic Metal Cations in Synthetic Macrocyclic Hosts. J. Am. Chem. Soc.
1988, 110, 4994–5008. (b) Caneschi, A.; Sorace, L.; Casellato, U.; To-
masin, P.; Vigato, P. A. d- or f-Mononuclear and Related Heterodinuclear
Complexes with [1+1] Asymmetric Compartmental Macrocycles. Eur. J.
Inorg. Chem. 2004, 3887–3900. (c) Pierre, V. C.; Harris, S. M.; Pailloux, S.
L. Comparing Strategies in the Design of Responsive Contrast Agents for
Magnetic Resonance Imaging: A Case Study with Copper and Zinc. Acc.
Chem. Res. 2018, 51, 342–351.
(17) (a) Bertini, I.; Luchinat, C. NMR of Paramagnetic Molecules in Bio-
logical Systems; The Benjamin/Cummings Publishing Company, Inc.:
Menlo Park, 1986. (b) Bertini, I.; Luchinat, C. The Hyperfine Shift. Coord.
Chem. Rev. 1996, 150, 29–75. (c) Bertini, I.; Luchinat, C. Relaxation.
Coord. Chem. Rev. 1996, 150, 77–110. (d) Bertini, I.; Luchinat, C.; Parigi,
G. Solution NMR of Paramagnetic Molecules: Applications to Metallobio-
molecules and Models; Elsevier Science B.V.: Amsterdam, 2001.
(18) Reath, A. H.; Ziller, J. W.; Tsay, C.; Ryan, A. J.; Yang, J. Y. Redox
Potential and Electronic Structure Effects of Proximal Nonredox Active
Cations in Cobalt Schiff Base Complexes. Inorg. Chem. 2017, 56, 3713–
3718.
(11) (a) Liu, G.; Li, Y.; Sheth, V. R.; Pagel, M. D. Imaging in Vivo Extra-
cellular pH with a Single Paramagnetic Chemical Exchange Saturation
Transfer Magnetic Resonance Imaging Contrast Agent. Mol. Imaging 2012,
11, 47–57. (b) Sheth, V. R.; Li, Y.; Chen, L. Q.; Howison, C. M.; Flask, C.
A.; Pagel, M. D. Measuring in Vivo Tumor pHe with CEST-FISP MRI.
Magn. Reson. Med. 2012, 67, 760–768. (c) McVicar, N.; Li, A. X.; Suchý,
M.; Hudson, R. H. E.; Menon, R. S.; Bartha, R. Simultaneous in Vivo pH
and Temperature Mapping Using a PARACEST-MRI Contrast Agent.
Magn. Reson. Med. 2013, 70, 1016–1025. (d) Delli Castelli, D.; Ferrauto,
G.; Cutrin, J. C.; Terreno, E.; Aime, S. In Vivo Maps of Extracellular pH in
Murine Melanoma by CEST-MRI. Magn. Reson. Med. 2014, 71, 326–332.
(e) Rancan, G.; Delli Castelli, D.; Aime, S. MRI CEST at 1T with Large µeff
Ln3+ Complexes Tm3+-HPDO3A: An Efficient MRI pH Reporter. Magn.
Reson. Med. 2016, 75, 329–336. (f) Wu, Y.; Zhang, S.; Soesbe, T. C.; Yu,
J.; Vinogradov, E.; Lenkinski, R. E.; Sherry, A. D. pH Imaging of Mouse
Kidneys in Vivo Using a Frequency-Dependent ParaCEST Agent. Magn.
Reson. Med. 2016, 75, 2432–2441. (g) Tsitovich, P. B.; Cox, J. M.; Spern-
yak, J. A.; Morrow, J. R. Gear Up for a pH Shift: A Responsive Iron(II) 2-
Amino-6-Picolyl-Appended Macrocyclic ParaCEST Agent That Protonates
at a Pendent Group. Inorg. Chem. 2016, 55, 12001–12010. (h) Thora-
rinsdottir, A. E.; Du, K.; Collins, J. H. P.; Harris, T. D. Ratiometric pH Im-
(19) APEX2, version. 2009.1; Bruker Analytical X-Ray Systems, Inc.: Mad-
ison, WI, 2009.
(20) Sheldrick, G. M. SADABS, version 2.03; Bruker Analytical X-ray Sys-
tems, Inc.: Madison, WI, 2000.
(21) (a) Sheldrick, G. M. SHELXTL, version 6.12; Bruker Analytical X-ray
Systems, Inc.: Madison, WI, 2000. (b) Sheldrick, G. M. SHELXT–
Integrated Space-Group and Crystal-Structure Determination. Acta Crystal-
logr., Sect. A: Found. Adv. 2015, 71, 3–8. (c) Dolomanov, O. V.; Bourhis,
L. J.; Gildea, R. J.; Howard, J. A. K.; Puschmann, H. OLEX2: A Complete
Structure Solution, Refinement and Analysis Program. J. Appl. Crystallogr.
2009, 42, 339–341.
(22) Chilton, N. F.; Anderson, R. P.; Turner, L. D.; Soncini, A.; Murray, K.
S. PHI: A Powerful New Program for the Analysis of Anisotropic Mono-
meric and Exchange-Coupled Polynuclear d- and f-Block Complexes. J.
Comput. Chem. 2013, 34, 1164–1175.
(23) (a) Evans, D. F. The Determination of the Paramagnetic Susceptibility
of Substances in Solution by Nuclear Magnetic Resonance. J. Chem. Soc.
1959, 0, 2003–2005. (b) Schubert, E. M. Utilizing the Evans Method with
a Superconducting NMR Spectrometer in the Undergraduate Laboratory. J.
Chem. Educ. 1992, 69, 62.
(24) Bain, G. A.; Berry, J. F. Diamagnetic Corrections and Pascal’s Con-
stants. J. Chem. Educ. 2008, 85, 532–536.
(25) Dixon, W. T.; Ren, J.; Lubag, A. J. M.; Ratnakar, J.; Vinogradov, E.;
Hancu, I.; Lenkinski, R. E.; Sherry, A. D. A Concentration-Independent
Method to Measure Exchange Rates in PARACEST Agents. Magn. Reson.
Med. 2010, 63, 625–632.
(26) (a) Hirose, K. A Practical Guide for the Determination of Binding Con-
stants. J. Incl. Phenom. Macrocycl. Chem. 2001, 39, 193–209. (b) Ellis, A.
L.; Mason, J. C.; Lee, H.-W.; Strekowski, L.; Patonay, G.; Choi, H.; Yang,
J. J. Design, Synthesis, and Characterization of a Calcium-Sensitive Near
Infrared Dye. Talanta 2002, 56, 1099–1107. (c) Reuter, K.; Buchner, M.
aging with a CoII MRI Probe via CEST Effects of Opposing pH Depend-
2
ences. J. Am. Chem. Soc. 2017, 139, 15836–15847. (i) Srivastava, K.; Fer-
rauto, G.; Young, V. G.; Aime, S.; Pierre, V. C. Eight-Coordinate, Stable
Fe(II) Complex as a Dual 19F and CEST Contrast Agent for Ratiometric pH
Imaging. Inorg. Chem. 2017, 56, 12206–12213. (j) Thorarinsdottir, A. E.;
Tatro, S. M.; Harris, T. D. Electronic Effects of Ligand Substitution in a
Family of CoII PARACEST pH Probes. Inorg. Chem. 2018, 57, 11252–
2
11263. (k) Thorarinsdottir, A. E.; Harris, T. D. Dramatic Enhancement in
pH Sensitivity and Signal Intensity through Ligand Modification of a Dico-
balt PARACEST Probe. Chem. Commun. 2019, 55, 794–797.
(12) (a) Zhang, S.; Malloy, C. R.; Sherry, A. D. MRI Thermometry Based
on PARACEST Agents. J. Am. Chem. Soc. 2005, 127, 17572–17573. (b)
Li, A. X.; Wojciechowski, F.; Suchý, M.; Jones, C. K.; Hudson, R. H. E.;
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
R.; Thiele, G.; von Hänisch, C. Stable Alkali-Metal Complexes of Hybrid
(30) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Thermody-
Disila-Crown Ethers. Inorg. Chem. 2016, 55, 4441–4447. (d) Pauric, A. D.;
Jin, S.; Fuller, T. J.; Balogh, M. P.; Halalay, I. C.; Goward, G. R. NMR
Determination of the Relative Binding Affinity of Crown Ethers for Man-
ganese Cations in Aprotic Nonaqueous Lithium Electrolyte Solutions. J.
Phys. Chem. C 2016, 120, 3677–3683.
(27) Kuzmič, P. Program DYNAFIT for the Analysis of Enzyme Kinetic
Data: Application to HIV Proteinase. Anal. Biochem. 1996, 237, 260–273.
(28) (a) Wood, P. M. The Potential Diagram for Oxygen at pH 7. Biochem.
J. 1988, 253, 287−289. (b) Sawyer, D. T.; Sobkowiak, A.; Roberts, J. L.
Electrochemistry for Chemists, 2nd ed.; John Wiley & Sons: New York,
1995.
namic and Kinetic Data for Macrocycle Interactions with Cations and Ani-
ons. Chem. Rev. 1991, 91, 1721–2085.
1
2
3
4
5
6
7
8
(31) (a) Shannon, R. D. Revised Effective Ionic Radii and Systematic Stud-
ies of Interatomic Distances in Halides and Chalcogenides. Acta Crystal-
logr., Sect. A 1976, 32, 751–767. (b) Lang, P. F.; Smith, B. C. Ionic Radii
for Group 1 and Group 2 Halide, Hydride, Fluoride, Oxide, Sulfide, Sele-
nide and Telluride Crystals. Dalton Trans. 2010, 39, 7786–7791.
(32) Drew, M. G. B.; Harding, C. J.; McKee, V.; Morgan, G. G.; Nelson, J.
Geometric Control of Manganese Redox State. J. Chem. Soc., Chem. Com-
mun. 1995, 0, 1035–1038.
9
(29) Chantarojsiri, T.; Ziller, J. W.; Yang, J. Y. Incorporation of Redox-
Inactive Cations Promotes Iron Catalyzed Aerobic C–H Oxidation at Mild
Potentials. Chem. Sci. 2018, 9, 2567–2574.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
ACS Paragon Plus Environment

Page 11 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment