Paracetamol and other acetanilide analogs as inter-molecular hydrogen bonding assisted diamagnetic CEST MRI contrast agents

Article abstract of DOI:10.1039/d0ra10410h

Paracetamol and a few other acetanilide derivatives are reported as a special class of diamagnetic Chemical Exchange Saturation Transfer (diaCEST) MRI contrast agents, that exhibit contrast only when the molecules form inter-molecular hydrogen bonding mediated molecular chains or sheets. Without the protection of the hydrogen bonding their contrast producing labile proton exchanges too quickly with the solvent to produce any appreciable contrast. Through a number of variable temperature experiments we demonstrate that under the conditions when the hydrogen bond network breaks and the high exchange returns back, the contrast drops quickly. The well-known analgesic drug paracetamol shows 12% contrast at a concentration of 15 mM at physiological conditions. With the proven safety track-record for human consumption and appreciable physiological contrast, paracetamol shows promise as a diaCEST agent forin vivostudies.

Full text of DOI:10.1039/d0ra10410h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Paracetamol and other acetanilide analogs as inter-
molecular hydrogen bonding assisted diamagnetic
CEST MRI contrast agents†
Cite this: RSC Adv., 2021, 11, 6526
*
and Arindam Ghosh
*
Subhayan Chakraborty,
S. Peruncheralathan
Paracetamol and a few other acetanilide derivatives are reported as a special class of diamagnetic Chemical
Exchange Saturation Transfer (diaCEST) MRI contrast agents, that exhibit contrast only when the molecules
form inter-molecular hydrogen bonding mediated molecular chains or sheets. Without the protection of
the hydrogen bonding their contrast producing labile proton exchanges too quickly with the solvent to
produce any appreciable contrast. Through
a number of variable temperature experiments we
demonstrate that under the conditions when the hydrogen bond network breaks and the high exchange
returns back, the contrast drops quickly. The well-known analgesic drug paracetamol shows 12%
contrast at a concentration of 15 mM at physiological conditions. With the proven safety track-record for
human consumption and appreciable physiological contrast, paracetamol shows promise as a diaCEST
agent for in vivo studies.
Received 10th December 2020
Accepted 29th January 2021
DOI: 10.1039/d0ra10410h
rsc.li/rsc-advances
transfer the saturation through multiple exchange events with
the solvent during the long continuous irradiation. It is quite
1. Introduction
obvious that the CEST efficiency improves with increasing
exchange rate (kex) of a labile proton.7 However, if the exchange
becomes too fast on the NMR time scale, that is if kex becomes
larger than the offset (Du) of resonance frequency of a labile
proton from the bulk water, then the CEST efficiency starts
dropping again. For optimum CEST effect, the exchange rate
should be as high as possible as long as it remains in the slow to
medium exchange regime (kex # Du) to avoid coalescence.7a
Naturally, constant efforts are being made in the search of
molecules having labile protons with Du as large as possible, so
that faster-exchanging protons can take part in the CEST to
improve the CEST efficiency.
A class of CEST agents known as the paraCEST8 agents,
achieves Du of the order of several kilohertz through para-
magnetic downeld shi in the presence of a metal center.
However, the presence of a metal center raises safety concerns.9
Diamagnetic CEST agents10 (diaCEST), on the other hand, are
considered comparatively safer in absence of any metal center
but produces a much lower Du (typically less than 5 ppm) and
hence inferior CEST efficiency. Recently, two classes of diaCEST
agents were reported that produce comparatively larger offsets
and hence much higher CEST contrast by utilizing, respectively,
intramolecular hydrogen bonding and diatropic ring currents.
Some benzoic acid derivatives (salicylic11 and anthranilic12
acids, in particular), phenols13 and modied imidazole14
compounds belong to the rst class of molecules in which
intramolecular hydrogen bonding (will be referred as intraMHB
hereon for brevity) shis the labile proton responsible for CEST
contrast downeld (as large as 9.3 ppm for salicylic acid).
In any diagnostic imaging technique such as Magnetic Reso-
nance Imaging (MRI)1 or Computed Tomography (CT)2 it is
always desired that the region of interest remains at a different
contrast level from the surroundings. In MRI, differential water
proton density or water content is the source of natural contrast
that becomes useful in studying, for example, bones embedded
in so tissues. Also, in T1 or T2 weighted MRI imaging, the
relaxation time constants of water produce natural contrast.
However, in the absence of such natural contrast due to similar
water content of the study region and the surroundings or due
to unfavorable relaxation rates, articial contrast is achieved
through chemical MRI contrast agents3 following the seminal
work by Young et al.4 Traditional paramagnetic metal-
containing contrast agents bring contrast by altering either
the longitudinal or the transverse relaxation rate constants of
the water molecules.3a,5 In the year 2000, Ward et al. introduced
a new class of compounds, that produce contrast by partially
saturating the water through Chemical Exchange Saturation
Transfer (CEST).6 In particular, the CEST contrast agents
contain one or more labile protons that are saturated using low
power continuous irradiation immediately prior to the MRI
signal detection. The saturated labile protons then, in turn,
School of Chemical Sciences, National Institute of Science Education and Research
(NISER), HBNI, Bhubaneswar 752050, Odisha, India. E-mail: peru@niser.ac.in;
aringh@niser.ac.in
† Electronic supplementary information (ESI) available: 1H and 13C spectra of all
the compounds. Additional CEST z-spectra. Additional omega plots for exchange
rate calculation. See DOI: 10.1039/d0ra10410h
6526 | RSC Adv., 2021, 11, 6526–6534
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Table 1 Frequency offset, CEST contrast and exchange rates for
acetanilide derivatives. All experiments were performed at 15 mM
sample concentration in 10 mM PBS at the physiological condition of
Certain meso-substituted porphyrin derivatives, on the other
hand, form the second class of diaCEST agents in which the
diatropic ring current produced by the conjugated p-electrons
brings a large upeld shi (ꢀ8 ppm) to the labile inner-core
amine protons.15
pH 7.4 and 37 ꢂ
C
Compound
Du [ppm]
CEST contrast
12.4%
kex (ksꢃ1
)
Herein, we introduce a few amide (acetanilide, in particular)
derivatives as a new class of diaCEST agents, that produces
comparable contrast as the previous two classes in spite of the
fact that they do not show a similar large shi. Amide is already
a popular choice12,16 for diaCEST agents as amides are found in
various natural products, drugs and biomolecules and hence is
considered in general non-toxic.17 The amide moiety however
plays an additional role as the core structural motif in the
present work. Amides are known to form intra as well inter-
molecular hydrogen bonds.18 The compounds we present here
as diaCEST agents do not exhibit contrast as an individual entity
due to very fast exchange of their labile protons but do so only
when they form ‘intermolecular hydrogen bonding (interMHB
for brevity)’ mediated molecular chains or sheets. The
interMHB helps to bring the exchange rate down to a favorable
slow to medium exchange regime. In other words, the mole-
cules act as a CEST contrast agent only as a group. We show this
interesting behavior through a number of variable temperature
experiments. We demonstrate that under mild heating above
the physiological temperature when the interMHB chains start
breaking, the isolated molecules cease to contribute to observ-
able contrast as then the exchange rate breaches the slow to
medium exchange regime (kex # Du). Although the weaker
intrerMHB cannot produce as spectacular Du as is produced by
the much stronger intraMHB or the intense diatropic ring
current, it slows down the exchange rate very optimally. For
example, paracetamol (acetaminophen), a well-known analgesic
drug19 and a notable member of this proposed class of CEST
agents shows Du of only 4.43 ppm, but yet it produces nearly
similar contrast efficiency as compared to that shown by sali-
cylic acid. The interMHB plays here the crucial role of opti-
mizing the interplay between Du and kex which is pivotal for any
good CEST MRI contrast agent.
5.21
1.84
5.27
5.07
5.06
5.53
11.9%
6.8%
8.2%
0.5%
2.7
1.7
1.6
1.07
4.41
4.61
7.5%
8.1%
3.08
4.18
4.60
4.43
10.5%
12%
2.1
1.2
2.2 Sample preparation and instrumentation
15 mM solutions of acetanilide and its derivatives were
prepared using 10 mM (1ꢁ) PBS buffer. For the pH dependent
studies seven samples were prepared with pH values ranging
between 6.8 and 8.1 by using 0.5 M HCl and 0.5 M NaOH
solutions. pH measurements were performed using a Hanna
HI2210 pH-meter. All experiments were performed at the
2. Experiments and methods
2.1 Synthesis and characterization
Acetanilide and substituted anilines were purchased from
Sigma-Aldrich and were used without further purication.
Other acetanilide derivatives (Table 1) were synthesized from
the respective commercially available substituted aniline
precursors following literature procedure.20 Acetic anhydride
(12.7 mmol) was added to the aniline precursors (10 mmol)
suspended in water (3 mL). The mixture was vigorously stirred
in a warm water bath. Once the solid particles got dissolved, the
reaction mixture was cooled. The solid product was then ltered
and washed with cold water. The solid acetyl derivatives were
nally recrystallized from hot water and dried at ambient
conditions. All the synthesized compounds were characterized
using 1H and 13C 1D NMR spectra acquired in DMSO-d6 as the
solvent.
ꢂ
physiological temperature of 37 C unless mentioned other-
wise. For CEST studies in the biological media, 15 mM
solutions at pH 7.4 were prepared by dissolving the necessary
compounds in 1 mL of either 4% Fetal Bovine Serum (FBS) or
4% Horse Serum (HS) solution. All NMR experiments were
performed on a 9.4 T Bruker AVANCE-III Nanobay NMR
spectrometer equipped with a BBFO broadband probe, except
for the exchange constant determination experiments.
Exchange constants were calculated from experiments
acquired on a 16.4 T Bruker AVANCE-III AV700 Ascend
spectrometer equipped with
a triple resonance room
temperature probe. D2O lled capillaries were placed
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 6526–6534 | 6527

View Article Online
RSC Advances
Paper
coaxially inside the NMR tube for deuterium lock. Methanol- Translational diffusion coefficient, D, was calculated from the
d4 was used for temperature calibration using standard negative slope (equal to Dz) of the linear t of ln(I/I0) plotted
procedure.
against g2.
2.3 Preparation of CEST z-spectra
2.6 Relaxivity measurements
CEST experiments were carried out with r.f. irradiation at
213 Hz (5 mT) eld strength for 3 s. Data were collected at 1 ppm
interval and normalized water peak intensity (100 ꢁ MZ/M0) was
plotted against irradiation offset frequency to obtain the CEST
z-spectra. MZ and M0 denote the water peak intensities,
respectively, with and without on-resonance (on the solute
exchangeable peak) saturation. CEST contrast or efficiency was
calculated from the asymmetric magnetization transfer ratio
(MTRassym) given by,7c
Relaxivity of a solute molecule is dened as the concentration
(expressed in mM in most situations) normalized change in
either longitudinal (R1) or transverse (R2) relaxation rate
constants of water in presence of the solute. Inversion recovery
method was used with 12 variable delays between 10 ms and 6 s
to measure R1. R2 was measured using Carr–Purcell–Meiboom–
Gill spin-echo sequence with 14 delays between 80 ms and
1.44 s. Sufficiently long recycle delay of 10 s was used for
accurate measurement of the relaxation rate constants. Relax-
ivities were calculated from the measured R1 and R2 values.
MðꢃDuÞ ꢃ MðDuÞ
CEST efficiency ¼
ꢁ 100%;
(1)
MðꢃDuÞ
M(Du) and M(ꢃDu) being, respectively, intensities of the water
peak aer on-resonance saturation (at the site of the
exchangeable peak, offset Du from water resonance frequency)
and off-resonance saturation (equal offset on the other side of
water peak, ꢃDu). In-house MATLAB (R2014b) scripts were used
for post-acquisition processing and plotting.
3. Results and discussion
3.1 Role of inter-molecular hydrogen bonding in producing
CEST contrast
At ambient temperatures, acetanilide (1) is known18 to form
molecular chains through interMHB as shown schematically in
Fig. 1a. The hydrogen bond removes the partial double bond
character of the C–N bond thereby facilitating a stable cong-
uration that relieves the steric interaction between the ortho and
the amine hydrogen atoms.18 We hypothesize that it is this
chain formation that makes acetanilide a diaCEST agent as the
interMHB lowers the exchange rate to the slow to medium
exchange window. At physiological serum condition of pH 7.4 at
37 ꢂC, acetanilide shows a signicant 12.4% contrast with
15 mM concentration (Fig. 1b). To conrm our hypothesis that
without the interMHB mediated molecular chains an individual
acetanilide would not show appreciable CEST, we performed
a series of variable temperature experiments. Assuming our
hypothesis true, we anticipated that at some temperature when
the interMHB chain would break, the CEST contrast would
simply disappear. Fig. 1c shows the stack-plot of the labile
amine proton of acetanilide acquired in DMSO-d6 as a function
of temperature. The gradual upeld shi of the NH proton with
temperature supports the slow reappearance of the electron
density that was originally withdrawn due to the formation of
the interMHB. When we plotted the peak shi (Dn) from the
position at 298 K, we found an interesting behavior (Fig. 1d). Up
to 313 K (40 ꢂC) the shi followed the expected linear rela-
tionship with temperature as generally is observed for many
2.4 Exchange rate calculation
Exchange rates were calculated using the method proposed by
Dixon et al.21 The method uses a computationally simple linear
regression method where the quantity MZ/(M0 ꢃ MZ) is tted
2
linearly against 1/u1 . u1 is related to the r.f. power B1 as u1 ¼
gB1, g being the gyromagnetic ratio of proton. The negative x-
axis intercept of the linear t gives the exchange rate as [1/u1-
2
]
¼ ꢃ1/(kex)2. The method gives the best result under the
intercept
condition that direct saturation (DS) of water is negligible,
saturation powers (in Hz) used are not larger than the offset Du
and saturation time is long enough to ensure complete satu-
ration. To satisfy these conditions, exchange rates were
measured at the highest magnetic eld available in the labo-
ratory, with long saturation duration and using a sinc pulse-
train that produces low DS. In the pulse train, 50 ms sinc pul-
ses were used with 4 ms inter-pulse delay. Maximum saturation
power was adjusted by taking the chemical shi offset of the
compound under study. All ttings were performed in Microso
Excel.
2.5 Diffusion measurements
Pulse eld Gradient Stimulated Echo (PGSE)22 method was solutes. However, beyond 313 K, the shi deviated from the
employed to determine the translational diffusion coefficient. linear relationship which strongly indicates the onset of an
To reduce the effect of Eddy current bipolar gradients were event, possibly the breaking up of the interMHB chain. The
used. Attenuation of all 1H NMR signals with a reasonable observed change in linewidth (LW) with temperature was found
signal to noise was measured as a function of gradient strength. to be even more interesting (Fig. 1e). For small molecules, the
Gradient strength was varied between 2% and 95% of the peak transverse relaxation rate constant generally drops with
gradient value of 48.15 gauss per cm. The normalized peak a decrease in rotational correlation time (with an increase in
intensity (I/I0) for any particular gradient strength (g) follows temperature). As a consequence, the linewidth should also drop
a relationship I/I0 ¼ exp(ꢃDzg2), I0 being the equilibrium as the sample heats up. Contrary to this expected behaviour the
intensity and z is a constant that depends on a number of linewidth was found to be monotonically increasing with
experimental parameters and can readily be calculated.22 temperature. Moreover, the linewidth also deviated from the
6528 | RSC Adv., 2021, 11, 6526–6534
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
Fig. 1 (a) Schematic representation of molecular chain formation through inter-molecular hydrogen bonding by acetanilide in solution. (b) The
CEST z-spectrum of acetanilide at the physiological condition (pH 7.4, 37 ꢂC). MTRassym peak on the x-axis shows 12.4% CEST contrast. (c) Stack
1D plot of the labile amine peak of acetanilide in DMSO-d6 as a function of temperature. Temperature dependent peak-shift with respect to the
position at 298 K (d) and peak line-width (e) of the acetanilide amine proton. (f) Exchange rate of the amine proton in water as a function of
temperature. The exchange rate crosses the important limit of Du (offset with respect to water peak position) at 333 K (g). (h) Natural logarithm of
the exchange constant against the inverse of temperature. (i) The ratio of effective hydrodynamic radii at temperature T and at 298 K. (j) CEST
contrast percentage as a function of exchange rate constant and temperature (inset). None of the dotted lines are mathematical fitting of data.
They simply help to visualize the pattern of the data points. Please refer to the main text for analysis of the data towards confirming the gradual
breaking up of the chain formation with increasing temperature.
expected linear behaviour (R2 changes nearly linearly with increase of kex at 333 K is possibly due to the complete disrup-
temperature for small molecules) beyond 313 K. This could only tion of the interMHB mediated molecular chain network. The
be explained if we assume that the acetanilide molecules jump discussion so far has picked up on a number of occasions the
between the free and the chain congurations causing exchange temperature 313 K as the onset temperature of an event. To
broadening. With an increase in temperature, the exchange further conrm that the event indeed was the breaking up of the
broadening should naturally increase. The amine proton of molecular chain, we estimated at different temperatures the
acetanilide in water shows expected non-linear increase in effective hydrodynamic radius (rT) of the molecules in the
solute–solvent exchange rate with temperature (Fig. 1f and S1– solution from the measured translational diffusion coefficient
S9, ESI† for corresponding z-spectra). At 333 K the increase in as per the following relationship,23
exchange rate shows an avalanche which takes the exchange
kBT
rT
¼
;
(2)
rate beyond the slow to medium exchange regime as can be seen
from the plot of (Du ꢃ kex) as a function of temperature (Fig. 1g).
When we tried to check if the non-linear variation followed an
Arrhenius type exponential equation and plotted ln(kex) as
a function of 1/T, we found that the plot deviated from the ex-
pected linear behaviour beyond 313 K just as happened for the
peak shi and the peak linewidth (Fig. 1h). The departure from
the linear behaviour clearly indicates that the rise of the
exchange rate was not entirely due to the rise of the solvent
temperature but also due to the gradual breaking of the
interMHB chain. Without the partial locking imposed by
interMHB, the labile proton exchanged too fast. The sudden
6phD
where, kB, T, h and D denote, respectively, the Boltzmann
constant, absolute solvent temperature, solvent viscosity at
temperature T and translational diffusion coefficient of acet-
anilide. In absence of any structural changes of the molecules,
the effective hydrodynamic radius should remain unchanged
and hence the ratio rT/rT for any two temperatures T and T0
0
should be equal to 1. Fig. 1i shows for different temperatures
the ratio r298/rT, which deviates from the expected value of 1
beyond 313 K as has been seen previously with other experi-
mental parameters. The fact that beyond 313 K, rT becomes
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 6526–6534 | 6529

View Article Online
RSC Advances
Paper
smaller than r298 conrms the breaking of the longer molecular anion at the basic pH. Thus the exchange rate as well as the
chains into smaller fragments. Finally, Fig. 1j shows CEST CEST percentage quite expectedly decreased considerably.
contrast efficiency of acetanilide as a function of the exchange Among the ortho substituents, o-methoxy (6) and o-hydroxy (7)
constant and the temperature (Fig. 1j inset). Fig. 1j shows acetanilide showed a drastic increase in kex most likely due to
a smooth initial increase in the CEST contrast as a function of the solvation of the molecules with the bulk water. Interestingly,
increasing kex (and also temperature) but beyond 313 K, the even with such a steep increase in kex, the CEST efficiency was
CEST efficiency starts abruptly dropping. A clear change in the found to be very moderate for both 6 and 7. In the case of
slope of the curves (the blue and red dotted guidelines) indicate hydroxy substitutions, both meta (8) and the para (9) substituted
once again the onset of the event of breaking of interMHB acetanilide derivatives produced a higher CEST effect than o-
chains. The population of the free acetanilide molecules in the hydroxy substitution, with 9 being the top of the lot in spite of
dynamic mixture cannot produce CEST contrast due to very fast the fact that 9 does not show the highest exchange rate among
exchange and as with increasing temperature the fractional the hydroxy-substituted acetanilide compounds. As CEST
population of free molecules increases, the CEST efficiency contrast depends on several other factors including relaxation
gradually drops. At 333 K when the free molecule population rate constants of the labile proton of the solute, and in the
dominates heavily, the CEST effect nearly disappears. The presence of interMHB network these parameters can widely
observation that the onset temperature (313 K) for all the events vary, we could not see any direct correlation of the exchange rate
is so close to the physiological temperature of 310 K made us and/or offset frequency with the CEST contrast efficiency of one
believe that without the interMHB, acetanilide would have had acetanilide derivative with respect to the other. Paracetamol (9)
an exchange rate at the physiological temperature too high for shows better CEST efficiency than most other derivatives in
any meaningful CEST contrast.
spite of the fact that it has one of the smallest frequency offsets.
3.2 Variation of exchange rate and CEST efficiency in
different acetanilide derivatives
3.3 Study of CEST properties of paracetamol
Aer exploring the electronic and steric inuences of different
substitutions on the CEST properties of acetanilide, we propose
para-hydroxyl acetanilide (also known popularly as para-
cetamol) as a potential CEST contrast agent. Paracetamol is
a widely used analgesic having a proven safety track record for
human consumption.24 It is easily synthesizable in pure form
from 4-aminophenol using the procedure discussed earlier. At
Upon being convinced that the acetanilide molecule as a single
entity is not effective and needs to form a colony through
interMHB to show contrast, we synthesized a number of water-
soluble derivatives of acetanilide to explore the system while
keeping the interMHB chain formation intact (Table 1 and
Fig. S10–S25, ESI†). The steric and electronic effects brought in
by different substitutions generally plays an important role in
determining the frequency offset, exchange rate and CEST
efficiency of a molecule. In the present scenario, the additional
inuence of interMHB was expected to play a pivotal role. In
presence of electron withdrawing groups like uorine or
carboxylic acid at the para position, the electron density around
the NH proton decreases resulting in a downeld shi of the
amide proton with respect to acetanilide. Contrary to this,
electron donating group at the para position increases the
electron density and causes an upeld shi of the amide
proton. In case of ortho substitutions, the eld effect produces
further upeld shi. For para hydroxy acetanilide the combi-
nation of inductive as well as the mesomeric effect of the
hydroxyl group is instrumental in producing maximum upeld
shi for the NH proton among all the derivatives studied. We
found that for all the para-substitutions, a negative correlation
exists between pKa of the molecule and the kex (Fig. S26–S32,
ESI† for corresponding z-spectra). In the presence of electron-
withdrawing substituents like uorine (2), the electron density
decreases at the nitrogen centre because of ꢃI effect, which in
turn increases the lability and hence kex of the amine proton.
However, we did not nd the CEST contrast increasing in Fig. 2 (a) The intermolecular hydrogen bond mediated molecular
sheet formation by paracetamol.23 (b) Overlaid z-spectra of 15 mM
accordance in comparison to 1. For para methyl (3) and para
methoxy (4) substituents both kex and the CEST efficiency
dropped from 1. In the case of carboxylic acid substitution at
the para-position (5) the electron density increases at the site of
solutions of paracetamol in 10 mM (1ꢁ) PBS buffer (green), Fetal
Bovine Serum (FBS, blue) and Horse Serum (HS, purple) at the physi-
ological condition acquired with 3 s saturation with 5 mT power. (c)
CEST efficiency (MTRassym) as a function of saturation duration at 5 mT
hydrogen bond donation due to the presence of carboxylate r.f. power.
6530 | RSC Adv., 2021, 11, 6526–6534
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
RSC Advances
ambient temperatures, paracetamol makes two-dimensional current study. The goodness of linear t is evident in all the
sheet through interMHB as shown in Fig. 2a.25 Both the plots. At the physiological condition, paracetamol shows an
hydroxyl and the carbonyl oxygen atoms take part in an alter- exchange rate of 1.26 ꢄ 0.04 ksꢃ1 which is well within the
native arrangement to form the sheet. Fig. 2b shows the overlaid desired slow to medium exchange window. Paracetamol shows
CEST z-spectra of paracetamol at physiological temperature of monotonically increasing exchange rate as the buffer pH moves
ꢂ
37 C and pH 7.4 in the buffer and two biological media, Fetal from acidic to basic (Fig. 3c). Such a pattern is well observed in
Bovine Serum (FBS) and Horse Serum (HS). At 15 mM concen- most CEST contrast agents. Following the thumb rule of CEST
tration, paracetamol produces 12% contrast in the buffer as well agents, the CEST efficiency also increases along with the
as in biological media with 3 s saturation at 5 mT irradiation exchange rate and reaches a maximum of 17.7% at pH 7.9
power. The CEST contrast increases with saturation duration (Fig. 3d). However, the CEST efficiency drops marginally at
(observed in buffer) and saturates at 15% above 6 s saturation a very basic pH of 8.1 due to exchange broadening.
duration (Fig. 2c).
Paracetamol has a proven safety record for human
Exchange rates between the two pools of protons, one consumption. But even then it is always desired that any foreign
constituted by the solute and the other by the solvent, dictates object be administered in the human body at a minimal
the efficiency of a CEST contrast agent. To ensure accurate possible concentration. To check the efficiency of paracetamol
measurement of exchange rates, all the experimental conditions as a CEST agent in the range of concentrations generally
required for exchange rate measurement using the method considered safe we recorded CEST spectra at six different
proposed by Dixon et al. were satised. Long 6 s saturation was concentrations ranging between 25 mM and 5 mM (Fig. 4a). The
used for complete saturation (as indicated by Fig. 2c), the CEST contrast was found dropping nearly linearly from
maximum power used was 639 Hz (corresponding to 15 mT eld approximately 17% (25 mM) to 7% (5 mM) in the said range
strength) which is well within Du ¼ 3276 Hz at 16.4 T (the (Fig. 4b).
highest magnetic eld available in the laboratory) and DS was
Finally, we calculated the relaxivities of paracetamol by
measured to be less than 3% even with the maximum applied r.f measuring concentration normalized changes in longitudinal
power. Fig. 3a shows the overlaid z-spectra of paracetamol at (for R1 relaxivity) and transverse (for R2 relaxivity) relaxation
different saturation elds at the physiological condition. The times of water in presence of paracetamol. The R1 and R2
corresponding linear relationship between MZ/(M0 ꢃ MZ) and 1/ relaxivities of paracetamol were found, respectively, 0.02 sꢃ1
u12 is shown in Fig. 3b for the physiological pH 7.4 and also for mMꢃ1 and 0.01 sꢃ1 mMꢃ1. The calculated value for R2 relaxivity
the two endpoints of the pH range (pH 7 and pH 8.1) for the matched a previous report where it has been discussed that the
high relaxivity is due to the phenolic proton of paracetamol.26
The phenolic proton executes a much faster exchange in
comparison to the CEST producing amine proton and hence
shows R2ex contrast instead. This makes paracetamol a very
Fig. 3 (a) Overlaid z-spectra of paracetamol at different irradiation
powers with 3 s saturation in the physiological condition. The peaks on
the x-axis show corresponding MTRassym. (b) The linear fit for MZ/(M0 ꢃ
2
MZ) plotted against 1/u1 for three pH values, 7, 7.4 and 8.1. The
goodness of fitting is given by the R2 values. Plots at other pH values Fig. 4 CEST contrast efficiency of paracetamol at different concen-
are given in Fig. S33–S39 of ESI.† Exchange rate constant (c) and CEST trations between 5 mM and 25 mM. (a) Overlaid z-spectra and (b)
efficiency (d) at different pH of the medium show gradual increase of MTRassym as a function of concentration. All experiments were per-
both at basic pH.
formed with 3 s saturation duration at 5 mT irradiation power.
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 6526–6534 | 6531

View Article Online
RSC Advances
Paper
special molecule in which two different labile protons produce DMSO) d 168.88, 150.00, 127.88, 124.65, 122.49, 120.61, 111.53,
MRI contrast following two completely different mechanisms.
56.05, 24.31.
5.6 N-(2-Hydroxyphenyl)acetamide (7)32
4. Conclusions
1H NMR (700 MHz, DMSO) d 9.77 (s, 1H), 9.31 (s, 1H), 7.66 (dd, J
¼ 8.0, 1.7 Hz, 1H), 7.01–6.91 (m, 1H), 6.86 (dd, J ¼ 8.0, 1.5 Hz,
1H), 6.76 (t, J ¼ 7.7 Hz, 1H), 2.09 (s, 3H). 13C NMR (176 MHz,
DMSO) d 169.54, 148.36, 126.82, 125.16, 122.88, 119.44, 116.37,
24.03.
In summary, we have demonstrated that the formation of
molecular chains or sheets through inter-molecular hydrogen
bonding can transform a collection of otherwise not suitable
acetanilide derivatives into diamagnetic CEST MRI contrast
agents. The hydrogen bonding network shis the contrast
producing labile protons downeld and lowers their exchange
constants so optimally that the performance of the contrast
5.7 N-(3-Hydroxyphenyl)acetamide (8)33
agents peaks just about at the physiological condition. A 1H NMR (700 MHz, DMSO) d 9.78 (s, 1H), 9.33 (s, 1H), 7.18 (d, J
notable member of the group is the well-known analgesic drug ¼ 2.1 Hz, 1H), 7.04 (t, J ¼ 8.0 Hz, 1H), 6.91 (dd, J ¼ 8.0, 2.0 Hz,
‘paracetamol’ which shows 12% contrast with 15 mM concen- 1H), 6.42 (dd, J ¼ 8.0, 2.4 Hz, 1H), 2.01 (s, 3H). 13C NMR (176
tration at the physiological temperature and pH. Paracetamol is MHz, DMSO) d 168.61, 158.01, 140.82, 129.73, 110.55, 110.18,
regularly administered both orally and through intravenous 106.59, 24.52.
route in humans and hence paracetamol has the potential of
being used as either oral contrast agent (OCA)27 or as extracel-
lular uid contrast agent (ECF).27 Nearly 95% of paracetamol
metabolism occurs in the liver and is generally considered safe
for human consumption.24 Although one can expect marginal
drop in CEST efficiency at the lower magnetic eld of clinical
MRI scanners, we believe our ndings make it a good candidate
nonetheless for in vivo applications as a diaCEST agent.
5.8 N-(4-Hydroxyphenyl)acetamide (9)34
1H NMR (700 MHz, DMSO) d 9.64 (s, 1H), 9.13 (s, 1H), 7.33 (d, J
¼ 8.8 Hz, 2H), 6.67 (d, J ¼ 8.7 Hz, 1H), 1.98 (s, 3H). 13C NMR (176
MHz, DMSO) d 167.96, 153.57, 131.50, 121.25, 115.4.
Author contributions
SC conceptualized the project. SP designed and synthesized the
compounds. SC collected and analysed data using in-house
MATLAB packages written by AG. Overall supervision of the
project was done by AG. Original dra was written by AG while
review and editing on nal dra was done by all.
5. Spectral characterization
5.1 N-(4-Fluorophenyl)acetamide (2)28
1H NMR (400 MHz, DMSO) d 9.97 (s, 1H), 7.60–7.56 (m, 2H), 7.12
(t, J ¼ 8.9 Hz, 1H), 2.03 (s, 3H). 13C NMR (176 MHz, DMSO)
d 168.60, 159.00, 157.64, 136.19, 121.24, 121.20, 115.66, 115.54,
24.28.
Conflicts of interest
There are no conicts to declare.
5.2 N-(4-Tolyl)acetamide (3)29
1H NMR (700 MHz, DMSO) d 9.78 (s, 1H), 7.45 (d, J ¼ 8.2 Hz,
2H), 7.08 (d, J ¼ 8.2 Hz, 2H), 2.24 (s, 3H), 2.02 (s, 3H). 13C NMR
(176 MHz, DMSO) d 168.50, 137.28, 129.45, 119.56, 24.35, 20.85.
Acknowledgements
The infrastructure provided by NISER and DAE, India is duly
acknowledged. SC thanks NISER, DAE for nancial support.
5.3 N-(4-Methoxyphenyl)acetamide (4)29
1H NMR (400 MHz, DMSO) d 9.76 (s, 1H), 7.47 (d, J ¼ 9.1 Hz,
1H), 6.86 (d, J ¼ 9.0 Hz, 1H), 3.71 (s, 3H), 2.00 (s, 3H). 13C NMR
(101 MHz, DMSO) d 168.18, 155.48, 132.99, 121.00, 114.25,
55.60, 24.25.
References
1 V. P. B. Grover, J. M. Tognarelli, M. M. E. Crossey, I. J. Cox,
S. D. Taylor-Robinson and M. J. W. McPhail, J. Clin. Exp.
Hepatol., 2015, 5, 246–255.
5.4 4-Acetamidobenzoic acid (5)30
2 E. Seeram, Radiol. Technol., 2018, 89, 279CT–302CT.
3 (a) Y.-D. Xiao, R. Paudel, J. Liu, C. Ma, Z.-S. Zhang and
S.-K. Zhou, Int. J. Mol. Med., 2016, 38, 1319–1326; (b)
C. F. G. C. Geraldes and S. Laurent, Contrast Media Mol.
Imaging, 2009, 4, 1–23.
4 I. R. Young, G. J. Clarke, D. R. Baffles, J. M. Pennock,
F. H. Doyle and G. M. Bydder, J. Comput. Tomogr., 1981, 5,
543–547.
1H NMR (400 MHz, DMSO) d 12.64 (s, 1H), 10.23 (s, 1H), 7.88 (d,
J ¼ 8.7 Hz, 2H), 7.69 (d, J ¼ 8.8 Hz, 1H), 2.08 (s, 3H). 13C NMR
(101 MHz, DMSO) d 169.33, 167.40, 143.80, 130.84, 125.36,
118.64, 24.62.
5.5 N-(2-Methoxyphenyl)acetamide (6)31
1H NMR (700 MHz, DMSO) d 9.11 (s, 1H), 7.92–7.91 (m, 1H),
7.09–7.04 (m, 1H), 7.03 (dd, J ¼ 8.2, 1.6 Hz, 1H), 6.89 (td, J ¼ 7.7,
1.6 Hz, 1H), 3.83 (s, 3H), 2.08 (s, 3H). 13C NMR (176 MHz,
5 (a) D. V. Hingorani, A. S. Bernstein and M. D. Pagel, Contrast
Media Mol. Imaging, 2015, 10, 245–265; (b) F. Ye, S. Laurent,
A. Fornara, L. Astol, J. Qin, A. Roch, A. Martini,
6532 | RSC Adv., 2021, 11, 6526–6534
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Paper
M. S. Toprak, R. N. Muller and M. Muhammed, Contrast
Media Mol. Imaging, 2012, 7, 460–468.
6 K. M. Ward, A. H. Aletras and R. S. Balaban, J. Magn. Reson.,
2000, 143, 79–87.
7 (a) A. D. Sherry and M. Woods, Annu. Rev. Biomed. Eng., 2008,
10, 391–411; (b) P. C. M. van Zijl and N. N. Yadav, Magn.
Reson. Med., 2011, 65, 927–948; (c) G. Liu, X. Song,
K. W. Y. Chan and M. T. McMahon, NMR Biomed., 2013,
26, 810–828.
8 (a) S. Aime, M. Botta, M. Fasano and E. Terreno, Chem. Soc.
Rev., 1998, 27, 19–29; (b) S. Aime, A. Barge, D. Delli Castelli,
F. Fedeli, A. Mortillaro, F. U. Nielsen and E. Terreno, Magn.
Reson. Med., 2002, 47, 639–648; (c) S. Aime, D. Delli Castelli
and E. Terreno, Angew. Chem., Int. Ed., 2002, 41, 4334–
4336; (d) S. Zhang, M. Merritt, D. E. Woessner,
R. E. Lenkinski and A. D. Sherry, Acc. Chem. Res., 2003, 36,
783–790; (e) M. Woods, D. E. Woessner and A. D. Sherry,
RSC Advances
F. Torrealdea, M. Rega, V. Rajkumar, S. P. Johnson,
S. Richardson, M. Gonçalves, H. G. Parkes, E. Arstad,
D. L. Thomas, R. B. Pedley, M. F. Lythgoe and X. Golay,
Nat. Med., 2013, 19, 1067–1072; (k) M. Haris, A. Singh,
K. Cai, K. Nath, R. Crescenzi, F. Kogan, H. Hariharan and
R. Reddy, J. Neurosci. Methods, 2013, 212, 87–93; (l) A. Bar-
Shir, G. Liu, K. W. Y. Chan, N. Oskolkov, X. Song,
N. N. Yadav, P. Walczak, M. T. McMahon, P. C. M. van Zijl,
J. W. M. Bulte and A. A. Gilad, ACS Chem. Biol., 2014, 9,
134–138; (m) K. W. Chan, T. Yu, Y. Qiao, Q. Liu, M. Yang,
H. Patel, G. Liu, K. W. Kinzler, B. Vogelstein, J. W. Bulte,
P. C. van Zijl, J. Hanes, S. Zhou and M. T. McMahon, J.
Controlled Release, 2014, 180, 51–59; (n) K. W. Y. Chan,
J. W. M. Bulte and M. T. McMahon, Wiley Interdiscip. Rev.:
Nanomed. Nanobiotechnol., 2014, 6, 111–124; (o)
S. Maruyama, J. Ueda, A. Kimura and K. Murase, Magn.
Reson. Med. Sci., 2016, 15, 324–334.
Chem. Soc. Rev., 2006, 35, 500–511; (f) S. Aime, 11 (a) X. Yang, X. Song, Y. Li, G. Liu, S. Ray Banerjee,
D. D. Castelli, S. G. Crich, E. Gianolio and E. Terreno, Acc.
Chem. Res., 2009, 42, 822–831; (g) S. Viswanathan,
Z. Kovacs, K. N. Green, S. J. Ratnakar and A. D. Sherry,
Chem. Rev., 2010, 110, 2960–3018; (h) I. Hancu,
W. T. Dixon, M. Woods, E. Vinogradov, A. D. Sherry and
M. G. Pomper and M. T. McMahon, Angew. Chem., Int. Ed.,
2013, 52, 8116–8119; (b) W. G. Lesniak, N. Oskolkov,
X. Song, B. Lal, X. Yang, M. Pomper, J. Laterra,
S. Nimmagadda and M. T. McMahon, Nano Lett., 2016, 16,
2248–2253.
R. E. Lenkinski, Acta Radiol., 2010, 51, 910–923; (i) 12 X. Song, X. Yang, S. Ray Banerjee, M. G. Pomper and
P. B. Tsitovich, J. M. Cox, J. A. Spernyak and J. R. Morrow,
Inorg. Chem., 2016, 55, 12001–12010; (j) P. J. Burns,
M. T. McMahon, Contrast Media Mol. Imaging, 2015, 10,
74–80.
J. M. Cox and J. R. Morrow, Inorg. Chem., 2017, 56, 4545– 13 X. Yang, N. N. Yadav, X. Song, S. Ray Banerjee, H. Edelman,
4554; (k) R. N. Pradhan, S. Chakraborty, P. Bharti,
J. Kumar, A. Ghosh and A. K. Singh, Dalton Trans., 2019,
48, 8899–8910.
9 C. Rydahl, H. S. Thomsen and P. Marckmann, Invest. Radiol.,
2008, 43, 141–144.
I. Minn, P. C. M. van Zijl, M. G. Pomper and M. T. McMahon,
Chem.–Eur. J., 2014, 20, 15824–15832.
14 X. Yang, X. Song, S. Ray Banerjee, Y. Li, Y. Byun, G. Liu,
Z. M. Bhujwalla, M. G. Pomper and M. T. McMahon,
Contrast Media Mol. Imaging, 2016, 11, 304–312.
10 (a) P. C. van Zijl, C. K. Jones, J. Ren, C. R. Malloy and 15 (a) X. Zhang, Y. Yuan, S. Li, Q. Zeng, Q. Guo, N. Liu, M. Yang,
A. D. Sherry, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 4359–
4364; (b) W. Ling, R. R. Regatte, G. Navon and A. Jerschow,
Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 2266; (c) M. Haris,
K. Cai, A. Singh, H. Hariharan and R. Reddy, NeuroImage,
Y. Yang, M. Liu, M. T. McMahon and X. Zhou, Magn. Reson.
Med., 2019, 82, 577–585; (b) S. Chakraborty, M. Das,
A. Srinivasan and A. Ghosh, New J. Chem., 2021, 45, 1262–
1268.
2011, 54, 2079–2085; (d) K. Cai, M. Haris, A. Singh, 16 (a) S. Aime, L. Calabi, L. Biondi, M. De Miranda, S. Ghelli,
F. Kogan, J. H. Greenberg, H. Hariharan, J. A. Detre and
R. Reddy, Nat. Med., 2012, 18, 302–306; (e) G. Liu,
M. Moake, Y.-e. Har-el, C. M. Long, K. W. Y. Chan,
A. Cardona, M. Jamil, P. Walczak, A. A. Gilad, G. Sgouros,
P. C. M. van Zijl, J. W. M. Bulte and M. T. McMahon,
Magn. Reson. Med., 2012, 67, 1106–1113; (f) K. W. Y. Chan,
M. T. McMahon, Y. Kato, G. Liu, J. W. M. Bulte,
Z. M. Bhujwalla, D. Artemov and P. C. M. van Zijl, Magn.
Reson. Med., 2012, 68, 1764–1773; (g) A. Bar-Shir, G. Liu,
M. M. Greenberg, J. W. M. Bulte and A. A. Gilad, Nat.
Protoc., 2013, 8, 2380–2391; (h) A. Bar-Shir, G. Liu,
L. Paleari, C. Rebaudengo and E. Terreno, Magn. Reson.
`
Med., 2005, 53, 830–834; (b) D. L. Longo, W. Dastru,
G. Digilio, J. Keupp, S. Langereis, S. Lanzardo, S. Prestigio,
O. Steinbach, E. Terreno, F. Uggeri and S. Aime, Magn.
Reson. Med., 2011, 65, 202–211; (c) D. L. Longo, P. Z. Sun,
L. Consolino, F. C. Michelotti, F. Uggeri and S. Aime, J.
Am. Chem. Soc., 2014, 136, 14333–14336; (d) S. Sinharay,
E. A. Randtke, C. M. Howison, N. A. Ignatenko and
M. D. Pagel, Mol. Imaging Biol., 2018, 20, 240–248; (e)
X. Cai, J. Zhang, J. Lu, L. Yi, Z. Han, S. Zhang, X. Yang and
G. Liu, Chem.–Eur. J., 2020, 26, 11705–11709.
Y. Liang, N. N. Yadav, M. T. McMahon, P. Walczak, 17 (a) A. Greenberg, C. M. Breneman and J. F. Liebman, The
S. Nimmagadda, M. G. Pomper, K. A. Tallman,
M. M. Greenberg, P. C. M. van Zijl, J. W. M. Bulte and
A. A. Gilad, J. Am. Chem. Soc., 2013, 135, 1617–1624; (i)
K. W. Y. Chan, G. Liu, X. Song, H. Kim, T. Yu, D. R. Arin,
A. A. Gilad, J. Hanes, P. Walczak, P. C. M. van Zijl,
J. W. M. Bulte and M. T. McMahon, Nat. Mater., 2013, 12,
268–275; (j) S. Walker-Samuel, R. Ramasawmy,
Amide Linkage: Structural Signicance in Chemistry,
Biochemistry, and Materials Science, Wiley, 2000; (b) J. Pitzer
and K. Steiner, J. Biotechnol., 2016, 235, 32–46; (c)
A. K. Agrahari, A. K. Singh, A. S. Singh, M. Singh, P. Maji,
S. Yadav, S. Rajkhowa, P. Prakash and V. K. Tiwari, New J.
Chem., 2020, 44, 19300–19313; (d) C. W. Evans, S. Edwards,
J. A. Kretzmann, G. L. Nealon, R. Singh, T. D. Clemons,
© 2021 The Author(s). Published by the Royal Society of Chemistry
RSC Adv., 2021, 11, 6526–6534 | 6533

View Article Online
RSC Advances
Paper
M. Norret, C. A. Boyer and K. S. Iyer, New J. Chem., 2020, 44, 25 G.-W. An, H. Zhang, X.-L. Cheng, Q.-L. Zhuo and Y.-C. Lv,
20013–20020; (e) S. Zhang, J. Li, J. Li, N. Du, D. Li, F. Li and Struct. Chem., 2008, 19, 613.
J. Man, RSC Adv., 2020, 10, 34308–34322; (f) L. Posada, 26 J. Zhang, Y. Li, S. Slania, N. N. Yadav, J. Liu, R. Wang,
D. Davyt and G. Serra, RSC Adv., 2020, 10, 43653–43659.
J. Zhang, M. G. Pomper, P. C. van Zijl, X. Yang and G. Liu,
18 J. Binoy, N. B. Prathima, C. Murali Krishna, C. Santhosh,
Chem.–Eur. J., 2018, 24, 1259–1263.
I. Hubert Joe and V. S. Jayakumar, Laser Phys., 2006, 16, 27 Y.-D. Xiao, R. Paudel, J. Liu, C. Ma, Z.-S. Zhang and
1253–1263.
S.-K. Zhou, Int. J. Mol. Med., 2016, 38, 1319–1326.
28 G. K. S. Prakash, M. D. Moran, T. Mathew and G. A. Olah, J.
Fluorine Chem., 2009, 130, 806–809.
19 B. J. Anderson, Pediatr. Anesth., 2008, 18, 915–921.
20 B. S. Furniss, A. J. Hannaford, P. W. G. Smith and
A. R. Tatchell, Vogel's textbook of practical organic chemistry, 29 S. R. Thopate, S. R. Kote, S. V. Rohokale and N. M. Thorat, J.
Longman Scientic & Technical, 5th edn, 1989, p. 985. Chem. Res., 2011, 35, 124–125.
21 W. T. Dixon, J. Ren, A. J. M. Lubag, J. Ratnakar, 30 S. Monch, M. Netzel, G. Netzel and M. Rychlik, Anal.
¨
E. Vinogradov, I. Hancu, R. E. Lenkinski and A. D. Sherry,
Magn. Reson. Med., 2010, 63, 625–632.
22 D. H. Wu, A. D. Chen and C. S. Johnson, J. Magn. Reson., Ser.
A, 1995, 115, 260–264.
Biochem., 2010, 398, 150–160.
31 B. P. Fors, K. Dooleweerdt, Q. Zeng and S. L. Buchwald,
Tetrahedron, 2009, 65, 6576–6583.
¨
32 R. Moreno-Corral, H. Hop, L. Machi-Lara and K. O. Lara,
23 C. C. Miller and J. Walker, Proc. R. Soc. London, Ser. A, 1924,
106, 724–749.
Eur. J. Org. Chem., 2011, 2148–2162, DOI: 10.1002/
ejoc.201001312.
24 (a) L. L. Mazaleuskaya, K. Sangkuhl, C. F. Thorn, 33 S. S. van Berkel, B. van der Lee, F. L. van Del and
G. A. FitzGerald, R. B. Altman and T. E. Klein,
F. P. J. T. Rutjes, Chem. Commun., 2009, 4272–4274, DOI:
Pharmacogenet. Genomics, 2015, 25, 416–426; (b)
10.1039/B906762K.
J. A. Forrest, J. A. Clements and L. F. Prescott, Clin. 34 A. S. El-Shahawy, S. M. Ahmed and N. K. Sayed, Spectrochim.
Pharmacokinet., 1982, 7, 93–107.
Acta, Part A, 2007, 66, 143–152.
6534 | RSC Adv., 2021, 11, 6526–6534
© 2021 The Author(s). Published by the Royal Society of Chemistry