A simple strategy for the visual detection and discrimination of Hg2+and CH3Hg+species using fluorescent nanoaggregates

Article abstract of DOI:10.1039/d1dt01455b

Fluorescent nanoaggregates (FNAs) based on phenanthroline-based amphiphiles show changes in solution color from colorless to yellow upon addition of both Hg²⁺(LOD ～4 ppb) and CH₃Hg⁺(LOD ～18 ppb). However, the extent of flu

Full text of DOI:10.1039/d1dt01455b

Dalton
Transactions
View Article Online
View Journal
PAPER
A simple strategy for the visual detection and
discrimination of Hg2+ and CH3Hg+ species
Cite this: DOI: 10.1039/d1dt01455b
using fluorescent nanoaggregates†
a,b
Nilanjan Dey
Fluorescent nanoaggregates (FNAs) based on phenanthroline-based amphiphiles show changes in solu-
tion color from colorless to yellow upon addition of both Hg2+ (LOD ∼4 ppb) and CH3Hg+ (LOD ∼18
ppb). However, the extent of fluorescence quenching is more prominent with Hg2+ (∼12 fold) than with
CH3Hg+ (∼4 fold). Also, unlike Hg2+, the interaction of CH3Hg+ needs more time, ∼10 min at room temp-
erature. Experimental evidence indicates that both mercury species coordinate with the phenanthroline
unit and facilitate the charge transfer interaction while destabilizing the nanoassembly. The lower charge
density on CH3Hg+ along with its large size compared to Hg2+ may be the reason for such observations.
Interestingly, FNAs show a selective response towards CH3Hg+ when pre-treated with EDTA. Further, ana-
lysis of heavy metal pollutants in drinking water and biological samples was performed. High recovery
values ranging from 96% to 103.0% were estimated along with relatively small standard deviations (<3%).
Low-cost, reusable test strips were designed for rapid, on-site detection of mercury species. Further, the
in situ formed metal complexes are allowed to interact with thiol-containing amino acids. As expected,
CH3Hg+, being less thiophillic, endures less interaction with cysteine. Mechanistic investigations indicate
that thiolated amino acids can bind with the metal ion center and form a tertiary complex (cooperative
interaction).
Received 3rd May 2021,
Accepted 27th May 2021
DOI: 10.1039/d1dt01455b
rsc.li/dalton
optical probes for CH3Hg+ are very few in number.4 However,
methylmercury compounds are known to be more toxic to
Introduction
Environmental pollution caused by heavy metal and metal- living organisms than the inorganic forms of mercury. Due to
loids, such as mercury, arsenic, cadmium, iron, copper, nickel, their superior lipid solubility, methylmercury species can
and lead (density >5 gm cm−3), has become a global concern easily cross biological membranes, including the blood–brain
due to their adverse effects on human health and well-being.1 barrier, and deposit in different internal organs, such as the
Although some of these metal ions play important roles in brain, kidneys and nerve system.5 Exposure to CH3Hg+ even at
several physiological processes and proper functioning of a lower concentration can cause prenatal brain damage, cogni-
enzymes, their excess accumulation in the body can cause tive and motion disorders, and vision and hearing loss, while
damage to one or more essential organs, resulting in severe excess intake results in Minamata disease.6 Owing to its less
health problems and even death.2 Lately, optical probes have thiophilic nature and lower charge compared to Hg2+, CH3Hg+
been involved extensively in the analysis of real-life samples species generally yield a slower response and require larger
due to their structural diversity, low maintenance cost, high amounts to receive a detectable optical signal.7
sensitivity, simple protocols, and quick response times.3
Additionally, it has been observed that depending upon
Among them, ratiometric colorimetric probes are of particular their size, charge density, coordination behavior and crystal
interest because they can detect toxic pollutants by the naked field stabilization energy, metal ions can bind with various
eye with little to no background interference. Although there biomolecules, such as nucleotides, amino acids, proteins,
are numerous reports on colorimetric sensing of Hg2+ ions, enzymes, and anions.8 For example, affinity of Ni2+ to histidine
has often been utilized for purifying histidine-tagged protein
molecules.9 Similarly, metal complexes based on Cu2+ or Zn2+
aGraduate School of Science, Kyoto University, Sakyo, Kyoto 606-8502, Japan.
are regularly used for selective extraction of phosphorylated
proteins and peptides.10
E-mail: nilanjan@hyderabad.bits-pilani.ac.in
bDepartment of Chemistry, BITS-Pilani Hyderabad Campus, Shameerpet,
Considering these facts, herein, we have reported the detec-
Hyderabad-500078, Telangana, India
tion as well as the differentiation of inorganic (Hg2+) and
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
organic mercury species (CH3Hg+) using fluorescent nanoag-
d1dt01455b
This journal is © The Royal Society of Chemistry 2021
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Scheme 1 Schematic showing differential response towards inorganic
(Hg2+) and organic (CH3Hg+) mercury species.
gregates (Scheme 1) composed of phenanthroline-based
amphiphiles. Both CH3Hg+ and Hg2+ could result in changes
of the solution color from colorless to yellow. However, unlike
Hg2+, CH3Hg+ needs a ∼10 min resting time. Also, the extent
of fluorescence quenching was found to be more prominent
for Hg2+ than for CH3Hg+. Interestingly, the compound could
show a specific interaction with CH3Hg+ when pre-treated with
EDTA. Further, we utilized the present system to analyze both
Hg2+ and CH3Hg+ species in natural water samples (important
for waste-water management), in the presence of plasma pro-
teins (important in clinical diagnosis), etc. Low-cost, reusable
paper strips were designed for rapid, on-site detection of poss-
ible contamination. Further, the in situ formed metal com-
plexes have been involved in probing sulfhydryl-containing
amino acids, such as cysteine and homocysteine. Mechanistic
investigations indicate that the amino acids can bind (coopera-
tively) to the metal complex via thiol units and destabilize the
charge-transfer state.
Fig. 1 (a) Schematic showing Hg2+/CH3Hg+-induced disassembly of
nanoaggregates of 1 (a TEM image of 1 is shown in the middle). (b) UV-
visible titration of 1 (10 µM) with Hg2+ at pH 7.0 in PBS buffer. (c)
Fluorescence titration of 1 (10 µM, λex: 320 nm) with Hg2+ at pH 7.0 in
PBS buffer.
spectra showed a very small hypsochromic shift even when
exposed to UV radiation for ∼30 min (365 nm, 300 mW cm−2).
This suggests fairly high photostability of the fluorescent
nanoaggregates (Fig. S3†).
Interaction with inorganic mercury at biological pH
In aqueous medium, compound 1 showed a change in solu-
tion color from colorless to yellow selectively in the presence
of Hg2+. As expected, the UV-visible spectrum showed the
appearance of a new red-shifted charge-transfer band at
∼420 nm upon the addition of Hg2+ ion. However, no change
in solution color or in the UV-visible spectra was observed
when other divalent or trivalent metal ions were added under
similar conditions (Fig. 2a). The concentration-variation
Results and discussion
Design and synthesis of the optical probe
Compound 1 was synthesized by following a literature-reported studies with Hg2+ ion showed a gradual increase in absorbance
protocol via the carbonyl-nucleophilic addition reaction of in the ∼420 nm region with a concomitant decrease at
1,10-phenanthroline-2,9-dicarbaldehyde with isoniazid. The ∼310 nm (ratiometric response) (Fig. 1c and S4†). The Job’s
compound could form a pH-dependent, thermosensitive nano plot analysis confirms 1 : 1 binding stoichiometry with an
assembly in aqueous medium.11 Thus, a broad, red-shifted inflection point at X = 0.5 (X = molar ratio of 1 and Hg2+)
fluorescence band (λmax = 465 nm, Φ = 0.14) was observed at (Fig. 2b). The titration studies also indicate that 1 can detect
pH 7.0 in PBS buffer. The presence of residual absorbance in a as low as 4.2 ppb Hg2+ in aqueous medium. The affinity con-
longer wavelength region in the UV-visible spectrum also sup- stant (log K) was estimated as 4.47 0.03 based on the 1 :
ported this assumption (Fig. S1†). Although nanoaggregates of 1 molecular model (Fig. S1b†).13 The formation of such
1 have already been reported, we independently proved the thermodynamically stable Hg2+-complexes with phenanthro-
assembly formation via dynamic light scattering and trans- line ligands has been reported earlier.14
mission electron microscopic studies (Fig. 1a and 3a).
pH-Variation studies revealed that the binding of 1 with
Considering the affinity of the phenanthroline moiety towards Hg2+ occurred effectively over a wide range of pH from 4.5 to 9
transition metal ions, we could anticipate that 1 would be a (Fig. 2c). We suspected that protonation at the phenanthroline
good candidate for analyzing heavy metal pollutants nitrogen ends below pH 4.5 might resist further interaction
(Fig. 1b).12 However, the presence of acyl hydrazone units on with Hg2+. Because real-life samples are often tainted by the
both sides of the phenanthroline unit is expected to modify presence of multiple toxic ions, we also examined the response
the metal ion binding efficiency of 1. Unlike in THF medium, of 1 towards Hg2+ ion in the presence of other competitive
compound 1 showed a broad, structureless excitation spec- metal ions (Fig. S5†). However, no significant change in
trum (at 454 nm) in PBS buffer (pH 7.0), presumably due to optical response was observed when Hg2+ ion was mixed with
the self-assembly formation (Fig. S2†). Also, the absorption other metal ions, even at excess concentrations.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
Fig. 3 (a) Hydrodynamic diameter (DH) of 1 (10 µM) upon addition of
Hg2+ (15 µM) at pH 7.0 in PBS buffer. (b) Fluorescence decay profile of 1
(10 µM, λex: 320 nm, λem: 465 nm) with Hg2+ (15 µM) at pH 7.0 in PBS
buffer. (c) Recovery experiment with EDTA showing the reversible inter-
action of 1 (10 µM) with Hg2+ (15 µM). (d) Comparison of the interactions
of 1 and 2 (10 µM) with Hg2+ at pH 7.0 in PBS buffer.
Fig. 2 (a) Change in absorbance of 1 (10 µM) upon addition of different
metal ions (15 µM) at pH 7.0 in PBS buffer. (b) Job’s plot analysis of 1
with Hg2+ at pH 7.0 in PBS buffer. (c) Effect of pH on the interaction of 1
(10 µM, λex: 320 nm) with Hg2+ (15 µM). (d) Effect of temperature on the
interaction of 1 (10 µM, λex: 320 nm) with Hg2+ (0–14 µM).
might conclude that in the present scenario, the Hg2+ ion is
binding to the phenanthroline site rather than the pyridine
units. The formation of the 1 : 1 complex was also evident
from the ESI-MS mass spectral analysis (Fig. S7†). To verify the
involvement of the pyridine end, we performed the experiment
with another compound (2) without pyridyl ends. The control
compound showed interaction with Hg2+, but to a lesser extent
(Fig. 3d). This apparent discrepancy can be attributed to the
differences in the electronic characteristics of the acylhydra-
zone units.
Rationalization of binding interaction
Under a long UV lamp, 1 showed cyan-colored FL, as evidenced
by the formation of a fluorescence band at ∼465 nm. Among
various transition metal ions, only Hg2+ ion induced a ∼15 nm
blue shift in the FL maximum along with ∼12-fold quenching
in intensity (Fig. 1d). These observations indicate the possi-
bility of Hg2+-induced dissociation of the preformed nano-
assembly (Fig. 3a).15 The high spin–orbit coupling constant of
Hg2+ could also be the contributing factor.16 To confirm this,
we performed dynamic light scattering studies both with and
Interaction with organic mercury species (CH3Hg+)
without Hg2+ ion. As anticipated, a reduction in the average When 1 was exposed to CH3Hg+, the formation of a yellow-
hydrodynamic diameter of the aggregates could be seen upon colored solution was noted. Here, we also witnessed the
the addition of Hg2+. On the other hand, temperature-depen- appearance of the CT band in the ∼410 nm region. However,
dent quenching studies with Hg2+ showed an increment in unlike Hg2+, here, the response time was considerably slow;
Stern–Volmer constant values at a higher temperature, indicat- ∼10 min incubation time was required (Fig. 4d). Titration
ing the dynamic nature of the quenching process (Fig. 2d). A experiments of 1 with CH3Hg+ were carried out under similar
reduction in average FL lifetime value was noticed when 1 was conditions to that of Hg2+. The UV-visible titration studies
treated with Hg2+, presumably due to dissociation of the aggre- showed that ∼7 equiv. CH3Hg+ (∼5-fold larger than Hg2+) were
gated structure (Fig. 3b). Because it is known in the literature needed for saturation of the optical signal (Fig. 4a). On the
that transition metal ions, such as Cu2+, Fe3+, and Hg2+, can other hand, the extent of fluorescence quenching was found to
either form a coordination complex or become involved in the be less prominent compared to inorganic mercury species, i.e.
hydrolysis of acyl hydrazone units.17 The EDTA-mediated Hg2+ (Fig. 4b). Unlike Hg2+, here, ∼4-fold quenching of the FL
recovery experiment suggests that the observed change in the intensity was observed upon addition of CH3Hg+. In this case,
UV-visible spectrum with Hg2+ is reversible, indicating no the Job’s plot analysis also indicates 1 : 1 binding stoichio-
possibility of an ion-induced hydrolysis reaction (Fig. 3c).
metry (Fig. S8a†). The affinity constant was calculated as 3.88
Further, the 1H-NMR titration of 1 with Hg2+ (0–1 equiv.) 0.02 by fitting the data with the Benesi–Hildebrand 1 : 1
was performed in DMSO-d6/D2O (1 : 1) mixture medium to binding equation (Fig. S8b†). Also, titration studies indicate
comment on the mode of binding (Fig. S6†). The titration that the minimum detectable concentration is 18.2 ppb for
studies reveal broadening of the NMR signals along with pro- CH3Hg+ species.
minent downfield shifts. The extents of the downfield shifts
Although the FL response of CH3Hg+ was found to be very
were more prominent for the protons adjacent to the imine similar to that of Hg2+, no change in absorbance was noticed
functional group and phenanthroline nitrogen ends. Thus, we when EDTA was added to the aqueous solution of 1 + CH3Hg+
This journal is © The Royal Society of Chemistry 2021
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 5 (a) Change in absorbance of 1 + Hg2+ (10 µM, 1 : 1) with different
amino acids (30 µM) at pH 7.0 in PBS buffer. (b) UV-visible titration of 1
+ Hg2+ (10 µM, 1 : 1) with cysteine at pH 7.0 in PBS buffer (the inset
shows images of vials with and without cysteine). (c) Interaction of 1 +
Hg2+ and 1 + CH3Hg+ with cysteine through cooperative binding.
Fig. 4 (a) UV-visible titration of 1 (10 µM) with CH3Hg+ at pH 7.0 in PBS
buffer. (b) Fluorescence titration of 1 (10 µM, λex: 320 nm) with CH3Hg+
at pH 7.0 in PBS buffer. (c) Changes in absorbance of 1 (10 µM) with
different metal ions (70 µM) in the presence of EDTA (0.1 mM). (d) Time-
dependent changes in absorbance with CH3Hg+ (70 µM) at pH 7.0 in
PBS buffer.
450 nm) (Fig. S11†). To verify whether the interaction of amino
acids (cysteine, homocysteine, etc.) is independent of Hg2+
binding, we performed a Hill plot analysis. The positive Hill
coefficient values indicate cooperative binding of amino acids
in both cases (Fig. 6a and S12†).19
(Fig. S9a†). Unlike Hg2+, EDTA was not able to remove CH3Hg+
from the complex. This is quite expected, as it is known that
the binding affinity of EDTA for CH3Hg+ is 1016 times weaker
than that for Hg2+ 18
Based on this observation, we could
.
develop a selective colorimetric assay for CH3Hg+ using a
mixture of 1 and EDTA as a probe. After pretreatment with
EDTA, no other metal ions (including Hg2+) except CH3Hg+
could induce the formation of the charge transfer band
(Fig. 4c). Thus, the present system can detect and distinguish
both organic (CH3Hg+) and inorganic mercury (Hg2+) species.
Rationalization of amino acid binding
Because probe 1 alone did not show any interaction with
either cysteine or homocysteine, it can be assumed that the
amino acids are interacting with the metal complex through
Hg2+ ion (Fig. S13a†). In general, amino acids can either form
a ternary adduct with an in situ formed metal complex or dis-
Tandem sensing of amino acids at biological pH
Considering the thiophilic nature of Hg2+ ion, we exploited the
in situ generated metal complex to analyze thiol-containing
amino acids, such as cysteine and homocysteine. To accom-
plish this, a mixture solution of 1 and Hg2+ in a 1 : 1 ratio was
freshly made and equilibrated for ∼30 min before the studies.
The addition of sulfhydryl-containing amino acids turned the
yellow solution colorless. The UV-visible spectral studies indi-
cate that the CT signal at ∼410 nm almost disappeared in the
presence of these amino acids with the formation of two new
bands at 280 and 352 nm, respectively (Fig. 5b and S9b, S10†).
In both cases, the titration reached saturation with the
addition of only 2 equivalents of the target amino acids. No
amino acids, other than cysteine and homocysteine, induced
detectable changes in the absorbance (Fig. 5a). A similar type
of spectral change was observed when an aqueous solution of
1 + CH3Hg+ was exposed to cysteine. However, the extent of the
change was less prominent than that of Hg2+. This may be due
to the poor thiophilicity of the CH3Hg+ species as compared to
Hg+ (Fig. 5c). On the other hand, addition of both cysteine and
Fig. 6 (a) Hill-plot analysis showing the cooperative interaction with
cysteine in water (pH 7.0). (b) Reversibility check by sequential addition
of Hg2+ (15 µM) and cysteine (30 µM) in water (pH 7.0). (c) Energy-mini-
mized structures of 1 and the tertiary complex with Hg2+ and cysteine.
(d) UV-visible titration of 1 (10 µM) with CH3Hg+ (0–50 µM) at pH 7.0 in
homocysteine to 1 + Hg2+ resulted in a turn-on FL response (at PBS buffer (with ∼0.1 mg mL−1 HSA).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
sociate the preformed metal complex by forming a thermo- examined water samples collected from three different
dynamically more stable alternative complex with metal ions.20 sources, a tap, a pool, and the Arabian Sea, and we determined
The sequential addition of Hg2+ and cysteine in the same solu- the presence of CH3Hg+ and Hg2+ ions using the current col-
tion showed no recovery of the spectrum of native 1, which orimetric method.23 Before the analysis, the spectra of 1 were
ruled out the possibility of a reversible interaction (displace- recorded in blank water samples to rule out the possibility of
ment approach) between the metal complex and cysteine background inference. Then, the changes in absorbance were
(Fig. 6b). Although the absorption spectrum of 1 + Hg2+ upon recorded for the samples spiked with Hg2+ (0–2 ppm) and
treatment with cysteine did not show a similar CT band to 1 CH3Hg+ (0–10 ppm) (Fig. S18†). In all the cases, the presence
alone, the UV-visible spectra appeared to be quite different. of Hg2+ and/or CH3Hg+ was also quantified using the atomic
These differences again dismiss the idea of the release of ‘free’ absorption spectroscopic technique. Standard equations, as
probe during the interaction with amino acids. The 1H NMR obtained from titration studies, were used for quantitative ana-
titration of 1 + Hg2+ with cysteine showed regeneration of the lysis. A comparison using the AAS method indicated that the
peaks, but at different positions with different splitting pat- recovery values mostly varied from 96% to 103%, whereas the
terns (Fig. S13b†). Thus, we may assume that the species gen- RSDs were found to be less than 4% (Fig. 7c). The small stan-
erated in the reaction medium is not free probe but a ternary dard deviation values with high percentage recovery indicate
complex with reduced charge transfer characteristics. The fairly good accuracy of the present method.
ESI-MS spectrum of 1 + Hg2+ in the presence of both cysteine
Low-cost paper strips for on-site detection. It is evident that
and homocysteine indicates the formation of
a
ternary the regular monitoring of heavy-metal pollutants, such as Hg2+
in drinking water and marketed food items, is important for
complex with 1 : 1 : 2 stoichiometry (Fig. S14 and S15†).
Density functional calculations using the B3LYP/6-31G* their controlled uptake. However, in most cases, people have
level of theory showed that in the energy-minimized structure no access or means to avail themselves of sophisticated
of 1 + Hg2+, the metal ion binds to the phenanthroline and analytical facilities for such diagnoses. Considering this,
imine nitrogen ends (Fig. S16†).21 On the other hand, the herein, we have developed test-strips (paper strips coated with
ternary composite [1 : 1 : 2 for compound 1 : Hg2+ : cysteine] compound 1) for rapid, on-site detection of Hg2+.24 The paper
with cysteine showed a tetrahedral coordination site for Hg2+ strips coated with 1 showed no detectable color in normal day-
ion, comprised of two nitrogen ends of the phenanthroline light but showed blue fluorescence under a UV torch. However,
unit and sulfhydryl groups of two coordinated cysteine mole- the test strips turned yellow in color when exposed to both
cules (Fig. 6c). The Hg–N bond lengths were found to be 2.08 Hg2+ and CH3Hg+-contaminated solutions. Similarly, a concen-
and 2.10 Å, while the Hg–S bond lengths were 2.49 and 2.47 Å, tration-dependent quenching of the blue fluorescence inten-
respectively.
sity was noticed under long UV light. However, unlike Hg2+,
the extent of fluorescence quenching was less with CH3Hg+
(Fig. 7a). Further, the mercury-mediated change in color (or
Application in real-life sample analysis
Analysis of biological samples. Determination of CH3Hg+ or fluorescence) was quantified using readily available image pro-
Hg2+ in blood or urine samples of hospitalized people is essen- cessing software, ImageJ (Fig. 7b & S19a†). Further, the selecti-
tial to confirm a diagnosis of poisoning or to assist in forensic vity experiment was performed by applying the other metal
investigation. However, these biological fluids mostly contain ions onto these test strips under similar conditions
serum albumin (HSA) protein in large excess.22 Thus, to (Fig. S19b†). The observations indicate that the present system
achieve a quantitative estimation of mercury species in biologi- can sense Hg2+ in real-life samples without facing significant
cal fluids, it is essential that the sensor molecule show inter- interference from other competing metal ions. Further, the
action with the analytes even in presence of excess HSA. To
confirm this, we repeated the titration of 1 with both Hg2+ and
CH3Hg+ in the presence of 0.1 mg mL−1 of HSA (Fig. 6d and
S17†). In both cases, we could observe concentration-depen-
dent changes in absorbance upon the addition of mercury.
When the changes in the absorbance (Abs394 nm/Abs292
)
nm
values were plotted against the concentration of added
CH3Hg+ or Hg2+, we could obtain linear plots. This finding
indicates that even in the presence of plasma proteins, quanti-
tative estimation of CH3Hg+ or Hg2+ ion is possible using the
present method.
Waste-water management. Poisoning by Hg2+ even at a low
level results in muscle weakness, skin irritation, breathing
problem, poor coordination, etc.23 Therefore, it is important to
devise a reliable, low-cost method to check the presence of
Hg2+ ions in drinking water samples, particularly beyond its
Fig. 7 (a) Detection of Hg2+ and CH3Hg+ (0–40 µM) using test trips
(paper strips coated with 1). Effect of Hg2+ under (i) normal daylight and
(ii) a UV lamp; (iii) effect of CH3Hg+ under a UV lamp. (b) Quantification
of FL color change upon addition of Hg2+ and CH3Hg+ using ImageJ
permissible levels (i.e. 2 ppb). Considering this, here, we have software. (c) Detection of Hg2+ and CH3Hg+ in different water samples.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans.

View Article Online
Paper
Dalton Transactions
interaction while destabilizing the preformed nanoaggregates.
However, 1 could show selective interaction with CH3Hg+
when pretreated with EDTA. Then, the probe molecule was uti-
lized to analyze heavy metal pollutants in drinking water and
biological samples. The high recovery values (96%–103%) with
relatively small standard deviation (<4%) values indicate good
accuracy of the present method. Low-cost test strips were
designed for rapid, on-site detection. Further, the in situ
formed metal complex was utilized for sensing of thiol-con-
taining amino acids, such as cysteine (LOD: 6 ppb) and homo-
cysteine (LOD: 5 ppb). Mechanistic investigations indicate that
the amino acids can form tertiary complexes with the pre-
formed metal complex (cooperative manner). The compound
was utilized for intracellular bioimaging of both Hg2+ and
CH3Hg+ species.
Fig. 8 Intracellular imaging of Hg2+ (0–20 µM) (top lane) and CH3Hg+
(0–20 µM) (down lane) in HeLa cells using compound 1 (10 µM).
Hg2+-treated paper discs were dipped into EDTA solution,
which could effectively leach out the coordinated Hg2+ ion,
and the paper discs became ready to be reused. Thus, one can
even use the same paper disc more than once for the detection
of Hg2+. Thus, we can confirm that the present system can also
report reversible sensing of Hg2+ on a solid surface.
Conflicts of interest
There are no conflicts to declare.
Detection of CH3Hg+ and Hg2+ under intracellular con-
ditions. Subsequently, compound 1 was employed for detec-
tion of Hg2+ and CH3Hg+ under intracellular conditions in
HeLa cells (cervical cancer cell line). To achieve this, we first
performed a cell-viability assay (using MTT as the indicator
dye) in the presence of 1 (Fig. S20†).25 No significant reduction
in the cell viability was observed even in the presence of
∼50 μM 1, which ensures that 1 can be utilized for the intra-
cellular imaging of mercury species. Dose-dependent kinetic
studies suggested that 10 μM of 1 for a 40 min incubation at
37 °C was sufficient for complete cellular uptake. As expected,
the cells stained with 1 showed blue-colored fluorescence.
Then, the cells (pretreated with 1) were exposed to both Hg2+
(20 µM) and CH3Hg+ (20 µM) (Fig. 8). Although quenching of
the fluorescence intensity was observed in both cases, the
extent of quenching was prominent when the cells were incu-
bated with CH3Hg+. These observations indicate that the com-
pound can not only detect Hg2+ inside living mammalian
cells, but is also able to discriminate between Hg2+ and
CH3Hg+.
Acknowledgements
The author thanks JSPS (Japan) for a postdoctoral fellowship
and BITS-Pilani, Hyderabad for the research initiation grant
(RIG) and other technical support. Also, author wish to
acknowledge Prof. S. Bhattacharya (Director, IACS) for kind
support and encouragement.
Notes and references
1 J. Briffa, E. Sinagra and R. Blundell, Heliyon, 2020, 6, e04691.
2 (a) K. P. Carter, A. M. Young and A. E. Palmer, Chem. Rev.,
2014, 114, 4564–4601; (b) J. Gu, F. Zhang, Z. Zheng, X. Li,
R. Deng, Z. Zhou, L. Ma, W. Liu and Q. Wang, Chin. Chem.
Lett., 2021, 32, 87–91.
3 (a) B. Kaur, N. Kaur and S. Kumar, Coord. Chem. Rev., 2018,
358, 13–69; (b) N. Dey and S. Bhattacharya, Chem. Rec.,
2016, 16, 1934–1949; (c) N. Dey and S. Bhattacharya, Chem.
– Asian J., 2020, 15, 1759–1779.
4 (a) Y. K. Yang, S. K. Ko, I. Shin and J. Tae, Org. Biomol.
Chem., 2009, 7, 4590–4593; (b) M. Q. Yang, C. P. Zhou,
Y. Chen, J. J. Li, C. H. Zeng and S. Zhong, Sens. Actuators, B,
2017, 248, 589–596; (c) S. L. Pan, K. Li, L. L. Li, M. Y. Li,
L. Shi, Y. H. Liu and X. Q. Yu, Chem. Commun., 2018, 54,
4955–4958.
5 (a) W. N. Wu, H. Wu, Y. Wang, X. J. Mao, B. Z. Liu,
X. L. Zhao, Z. Q. Xu, Y. C. Fan and Z. H. Xu, RSC Adv., 2018,
8, 5640–5646; (b) X. Kong, L. J. Hou, X. Shao, S. M. Shuang,
Y. Wang and C. Dong, Spectrochim. Acta, Part A, 2019, 208,
131–139.
Conclusions
Fluorescent nanoaggregates based on a phenanthroline-based
amphiphilic probe (1) have been utilized for naked-eye sensing
of heavy metal pollutants. Addition of both Hg2+ (LOD: 4.8
ppb) and CH3Hg+ (LOD: 18.2 ppb) induced ratiometric
changes in solution color from colorless to yellow. Although
quenching of fluorescence intensity was witnessed with both
the species, the extent of change was more prominent with
Hg2+ (∼12-fold) than with CH3Hg+ (∼4-fold). Also, unlike Hg2+,
CH3Hg+ needs a relatively long incubation period (∼10 min).
The larger size of CH3Hg+ (soft acid) with less charge density
might be the reason for this observation. Mechanistic investi-
gation indicates that both Hg2+ and CH3Hg+ bind with phe-
nanthroline nitrogen ends and facilitate the charge transfer
6 D. R. Wallace and E. Bates, J. Clin. Toxicol., 2015, 5,
1000273.
7 L. N. Neupane, J. Park, P. K. Mehta, E. T. Oh, H. J. Park and
K. H. Lee, Chem. Commun., 2020, 56, 2941–2944.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
8 (a) J. A. Thomas, Dalton Trans., 2011, 40, 12005–12016; 16 P. Mahato, S. Saha, P. Das, H. Agarwallad and A. Das, RSC
(b) Z. Liu, W. He and Z. Guo, Chem. Soc. Rev., 2013, 42, Adv., 2014, 4, 36140–36174.
1568–1600; (c) N. Dey, D. Biswakarma, A. Gulyani and 17 (a) W. N. Wu, H. Wu, Y. Wang, X. J. Mao, B. Z. Liu,
S. Bhattacharya, ACS Sustainable Chem. Eng., 2018, 6,
12807–12816.
9 (a) L. E. Valenti, C. P. D. Pauli and C. E. Giacomelli, J. Inorg.
Biochem., 2006, 100, 192–200; (b) G. Agarwal, R. R. Naik
and M. O. Stone, J. Am. Chem. Soc., 2003, 125, 7408–7412.
10 (a) P. Mateus and R. Delgado, Dalton Trans., 2020, 49,
X. L. Zhao, Z. Q. Xu, Y. C. Fan and Z. H. Xu, RSC Adv., 2018,
8, 5640–5646; (b) X. Kong, L. J. Hou, X. Shao, S. M. Shuang,
Y. Wang and C. Dong, Spectrochim. Acta, Part A, 2019, 208,
131–139.
18 S. Oh, J. Jeon, J. Jeong, J. Park, E. T. Oh, H. J. Park and
K. H. Lee, Anal. Chem., 2020, 92, 4917–4925.
17076–17092; (b) D. Kraskouskaya, M. Bancerz, H. S. Soor, 19 D. Wang, X. Zhang, C. He and C. Duan, Org. Biomol. Chem.,
J. E. Gardiner and P. T. Gunning, J. Am. Chem. Soc., 2014,
136, 1234–1237.
11 (a) N. Dey and S. Bhattacharya, Chem. – Eur. J., 2017, 23,
16547–16554; (b) N. Dey, J. Mol. Liq., 2021, 327, 114799.
2010, 8, 2923–2925.
20 (a) J. M. Jung, C. Kim and R. G. Harrison, Sens. Actuators,
B, 2018, 255, 2756–2763; (b) Y. W. Choi, J. J. Lee, G. R. You
and C. Kim, RSC Adv., 2015, 5, 38308–38315.
12 (a) N. Kaur, G. Kaur and P. Alreja, J. Photochem. Photobiol., A, 21 (a) N. Dey, A. Ali, M. Kamra and S. Bhattacharya, J. Mater.
2018, 353, 138–142; (b) N. Kaur and P. Alreja, Tetrahedron
Lett., 2015, 56, 182–186.
13 (a) N. Dey, S. Bhattacharjee and S. Bhattacharya,
Chem. B, 2019, 7, 986–993; (b) N. Dey, N. Kumari,
D. Bhagat and S. Bhattacharya, Tetrahedron, 2018, 74,
4457–4465.
ChemistrySelect, 2020, 5, 452–462; (b) N. Dey, N. Kumari, 22 B. M. W. Fong, T. S. Siu, J. S. K. Lee and S. Tam, J. Anal.
D. Biswakarma, S. Jha and S. Bhattacharya, Inorg. Chim.
Acta, 2019, 487, 50–57.
14 (a) J. H. Hu, J. B. Li, J. Qi and J. J. Chen, New J. Chem.,
2015, 39, 843–848; (b) H. Shuai, C. Xiang, L. Qian, F. Bin,
L. Xiaohui, D. Jipeng, Z. Chang, L. Jiahui and Z. Wenbin,
Toxicol., 2007, 31, 281–287.
23 (a) H. N. Kim, W. X. Ren, J. S. Kim and J. Yoon, Chem. Soc.
Rev., 2012, 41, 3210–3244; (b) G. Li, J. Wang, D. Li, S. Liu,
J. Yin, Z. Lai and G. Yang, Chin. Chem. Lett., 2021, 32,
1527–1531.
Dyes Pigm., 2021, 187, 109125; (c) Y. W. Sie, C. L. Li, 24 (a) N. Dey, Analyst, 2020, 145, 4937–4941; (b) N. Dey,
C. F. Wan, J. H. Chen, C. H. Hu and A. T. Wu, Inorg. Chim.
Acta, 2018, 469, 397–401.
15 N. Dey and S. Bhattacharya, Anal. Chem., 2017, 89,
10376–10383.
ChemistrySelect, 2020, 5, 6823–6827.
25 A. Chatterjee, M. Banerjee, D. G. Khandare, R. U. Gawas,
S. C. Mascarenhas, A. Ganguly, R. Gupta and H. Joshi,
Anal. Chem., 2017, 89, 12698–12704.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans.