Naphthalimide based fluorescent organic nanoparticles in selective sensing of Fe3+and as a diagnostic probe for Fe2+/Fe3+transition

Article abstract of DOI:10.1039/d0tb02450c

Fluorescent organic nanoparticles (FONPs) have attracted considerable attention as a practical and effective platform for sensing and imaging applications. The present article delineates the fabrication of FONPs derived from the naphthalimide based histidine appended amphiphile, NID. The self-assembly of NID in 99 vol% water in DMSO led to the formation of FONPs through J-type aggregation. Aggregation-induced emission (AIE) was observed due to the pre-associated excimer of NID with bluish green emission at 470 nm along with intramolecular charge transfer (ICT). The emission of NID FONPs was utilized for selective sensing of Fe3+ and bioimaging of Fe3+ inside mammalian cells. The fluorescence intensity of the FONPs was quenched with the gradual addition of Fe3+ due to the formation of a 1?:?1 stoichiometric complex with the histidine residue of NID. The morphology of the FONPs transformed from spherical to spindle upon the complex formation of NID with Fe3+. The limit of detection (LOD) of this AIE based turn-off chemosensor for Fe3+ was found to be 12.5 ± 1.2 μM having high selectivity over other metal ions. On the basis of the very low cytotoxicity and selective sensing of Fe3+, NID FONPs were successfully employed for bioimaging of Fe3+ ions through fluorescence quenching within mammalian cells (NIH3T3, B16F10). Considering the varying oxidative stress inside different cells, NID FONPs were used for detecting Fe2+ to Fe3+ redox state transition selectively inside cancer cells (B16F10) in comparison to non-cancerous cells (NIH3T3). Selective sensing of cancer cells was substantiated by co-culture experiment and flow cytometry. Hence, NID FONPs can be a selective diagnostic probe for cancer cells owing to their higher H2O2 content. This journal is

Full text of DOI:10.1039/d0tb02450c

View Article Online
View Journal
Journal of
Materials Chemistry B
Materials for biology and medicine
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: D. Sarkar, M.
Chowdhury and P. K. Das, J. Mater. Chem. B, 2020, DOI: 10.1039/D0TB02450C.
Volume
Number
21 January 2018
Pages 341-528
6
3
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry B
Materials for biology and medicine
rsc.li/materials-b
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2050-750X
PAPER
Wei Wei, Guanghui Ma et al.
Macrophage responses to the physical burden of cell-sized
particles
rsc.li/materials-b

Page 1 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
Naphthalimide based fluorescent organic nanoparticle in selective
sensing of Fe3+ and as diagnostic probe for Fe2+/Fe3+ transition
Deblina Sarkar, Monalisa Chowdhury and Prasanta Kumar Das*
School of Biological Sciences, Indian Association for the Cultivation of Science
Jadavpur, Kolkata – 700032, India.
*Corresponding Author: bcpkd@iacs.res.in
1

Journal of Materials Chemistry B
Page 2 of 45
View Article Online
DOI: 10.1039/D0TB02450C
Abstract
Fluorescent organic nanoparticles (FONPs) have attracted considerable attention as a practical
and effective platform for sensing and imaging applications. The present article delineates
fabrication of FONPs derived from naphthalimide based histidine appended amphiphile, NID.
The self-assembly of NID in 99 vol% water in DMSO led to the formation of FONPs through J-
type aggregation. Aggregation-induced emission (AIE) was observed due to the pre-associated
excimer of NID with bluish green emission at 470 nm along with intramolecular charge transfer
(ICT). The emission of NID FONPs was utilized for selective sensing of Fe3+ and bioimaging of
Fe3+ inside mammalian cells. The fluorescence intensity of the FONPs got quenched with
gradual addition of Fe3+ due to 1:1 stoichiometric complex formation with the histidine residue
of NID. Morphology of the FONPs transformed from spherical to spindle upon the complex
formation of NID with Fe3+. The limit of detection (LOD) of this AIE based turn-off
chemosensor for Fe3+ was found to be 12.5±1.2 M having high selectivity over other metal ions.
On the basis of the very low cytotoxicity and selective sensing of Fe3+, NID FONP was
successfully employed for bioimaging of Fe3+ ions through fluorescence quenching within
mammalian cells (NIH3T3, B16F10). Considering the varying oxidative stress inside different
cells, NID FONP was used for detecting Fe2+ to Fe3+ redox state transition selectively inside
cancer cells (B16F10) in comparison to non-cancerous cells (NIH3T3). Selective sensing of
cancer cells was substantiated by co-culture experiment and flow cytometry. Hence, NID FONP
can be a selective diagnostic probe for cancer cells owing to their higher H2O2 content.
2

Page 3 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
Introduction
Fluorescence sensing has been ever expanding research domain in many disciplines like
chemistry, biology, clinical biology, biotechnology, environmental science and many more.1-3
Fluorescence probes have significant advantages over other category of sensors for the detection
of environmentally and biologically relevant metal ions as it provides convenient, rapid and
sensitive detection of target analyte with high selectivity.4-7 In general, fluorophores suffer from
self-quenching because of the aggregation caused quenching (ACQ), low signal-to-noise ratio,
small Stokes’ shift and poor photostability.8 In 2001, Tang et al coined the concept of
aggregation-induced emission (AIE) through the development of a unique luminogen that
exhibited notably enhanced emission in the aggregated state.9 One of the primary reasons for the
AIE is restricted intramolecular rotation of the building block, which facilitates the conversion of
the excitation energy consumption from non-radiative relaxation to radiative decay.10-12 These
luminogens are generally either weakly emissive or non-emissive in solution phase. The
aggregation-induced enhancement in the emission by the luminogens (AIEgens) opens up new
potentials in sensing with advantages of accuracy, high sensitivity, photostability etc.13-15 These
FONPs with AIE property are interesting manifestation of self-assembly derived from π-
conjugated oligomers, polymers, or chromophores having variable functionality.16-19
The AIE active FONPs have been gaining attention owing to their ease of fabrication,
synthetic diversity, water solubility and cytocompatibility.16,20-22 Considering its high signal-to-
noise ratio, FONPs are finding notable significance in selective sensing of different analytes
including metal ions.21 Various transition metal ions have wide range of biological and
environmental relevance. Among all the transition metals, iron (Fe) is physiologically most
abundant and important in biological systems. It plays crucial roles in oxygen uptake, electron
3

Journal of Materials Chemistry B
Page 4 of 45
View Article Online
DOI: 10.1039/D0TB02450C
transfer, cellular metabolism, enzyme catalysis, transcriptional regulation and many other
processes in living systems.23-25 Nevertheless, both its deficiency and excess can induce a variety
of diseases in the human body.26-28 Iron has multiple oxidation states like Fe2+ or Fe3+ (or
transiently higher oxidative states) and it displays unique reactivity depending on the redox state.
Thus, detection of iron at its different redox states is essential for better understanding of its
physiological and pathological roles. Transition from Fe2+ to Fe3+ in living system is well known
phenomenon during generation of reactive oxygen species (ROS) upon interaction of Fe2+ with
hydrogen peroxide in Fenton reactions.29-31 Oxidative damage of cellular components owing to
the uncontrolled ROS production leads to a condition termed as oxidative stress.32-33 To diagnose
such situation, development of selective fluorescent sensors for Fe3+ and monitoring of Fe2+/Fe3+
transition in cellular system is highly warranted. Although detection of Fe3+ ion has attracted
notable attention,34-37 however reports on FONPs based sensing of Fe3+ with high sensitivity and
selectivity as well as monitoring of Fe2+ to Fe3+ transition inside living cells are really scarce.38,39
Herein, we synthesized naphthalimide based amphiphilic molecule (NID, Fig. 1) having
histidine substitution at the side chain as Fe3+ recognition unit. NID self-assembled to form
spherical aggregated FONPs in 99 vol% water in DMSO and exhibited AIE at 470 nm (bluish
green emission) through excimer formation together with intramolecular charge transfer. Among
the various metal ions, luminescence of NID FONPs selectively got quenched by Fe3+ with a
detection limit of 12.5±1.2 M. Notably, the complex formation of Fe3+ with NID led to the
transformation of morphology of the FONPs from spherical to spindle. Fe3+ ion sensing by NID
FONPs was also investigated inside mammalian cells through bioimaging. NID FONPs can
detect Fe2+ to Fe3+ transition selectively inside the cancer cells (B16F10) in comparison to non-
cancerous cell (NIH3T3) by exploiting higher content of H2O2 in cancer cells.
4

Page 5 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
Experimental section
Materials. N-(tert-butoxycarbonyl)-L-aspartic acid 4-benzyl ester, 1-naphthaleneethanol, 1,8-
naphthalic anhydride, thiazolyl blue tetrazolium bromide (MTT) and solvents for NMR were
procured
from
Sigma-Aldrich.
N,N-(dimethylamino)pyridine
(DMAP),
N,N-
dicyclohexylcarbodiimide (DCC), 4-N-hydroxybenzotriazole (HOBT), palladium 10% on
activated charcoal, trifluoroacetic acid (TFA), L-histidine, di-tert-butyl dicarbonate, ethylene
diamine, benzyl chloroformate, silica gel (100-200 mesh and 60-120 mesh), sodium sulfate
(Na2SO4), 2-aminopyridine, citric acid, L-Dopa, EDTA, folic acid, glycine and all organic
solvents, were bought from SRL, India. Dulbecco’s modified Eagle’s medium (DMEM), Fetal
bovine serum (FBS) and trypsin were purchased from Thermo Fisher scientific, India. We used
Milli-Q water for all the experiments. NMR spectra were recorded in AVANCE (Bruker)
spectrometer. Bruker Ultraflex MALDI-TOF spectrometer using 2,5-dihydroxy benzoic acid as
matrix. The specific rotation of the synthesized compound was measured in Anton Paar (MCP-
200) polarimeter.
Synthesis of NID. 1,8-naphthalimidebased L-histidine functionalized amphiphile was
synthesized by given pathway (Scheme S1, Electronic Supplementary Information (ESI)). At
first, coupling reaction between N-(tert-butoxycarbonyl)-L-aspartic acid 4-benzyl ester (1.2
equiv, 1 g) and 1-naphthaleneethanol (1.0 equiv, 0.425 g) was carried out using DCC (1.1 equiv,
0.561 g), HOBT (1.1 equiv, 0.367 g) DMAP (1.1 equiv, 0.332 g), in dry methylene chloride
(DCM). The reaction mixture was stirred for 12 h under nitrogen atmosphere. 1N HCl was used
to wash the organic part and consequently it was dried over anhydrous Na2SO4. The residue
obtained after evaporation was purified by column chromatography using 60-120 mesh silica gel
as the stationary phase and the eluent was 1% methanol in chloroform (yield = 80%, 0.97 g). The
5

Journal of Materials Chemistry B
Page 6 of 45
View Article Online
DOI: 10.1039/D0TB02450C
BOC group was deprotected in 5 h using TFA (1.5 equiv, 0.34 g) in dry DCM. After solvent
removal, the crude mass taken in ethyl acetate was twice washed with 10% aqueous Na2CO3
solution and water. Drying and evaporation of the solvent resulted in the formation of free amine
(yield = 81%, 0.626 g). 1,8-naphthalic anhydride (1.0 equiv, 0.288 g) and the free amine
(1.1equiv, 0.625 g) were refluxed (75 °C) in dry EtOH for 12 h. The residue obtained after
removal of EtOH was taken in DCM and washed with water. Purification of the condensed
product was carried out in 100-200 mesh silica gel using 3% MeOH/CHCl3 as eluent (yield =
72%, 0.6 g). Consequently, this product (0.6 g) was dissolved in dry THF and hydrogenated
using Pd on activated charcoal. The desired acid derivative was obtained after filtration and
evaporation of the solvent, THF (yield = 85%, 0.417 g). On the other hand, one end of ethylene
diamine (1 equiv, 0.25 g) was coupled with di-tert-butyl dicarbonate (1 equiv, 0.91 g) by drop
wise addition of di-tert-butyl dicarbonate in DCM at 0-5 C for 5 h. The solution was stirred for
8 h and washed with water and brine. Mono-BOC protected ethylenediamine was purified in
100-200 mesh silica gel chromatography using methanol-chloroform (5% v/v) as the eluent
(yield = 86%, 0.573 g). The previously prepared acid (1 equiv, 0.417 g) was conjugated with this
amine (1 equiv, 0.143 g) using DCC coupling method as mentioned above. The residue obtained
after working up, drying and evaporation was purified in 60-120 mesh silica gel chromatography
using 3% methanol in chloroform (v/v) as eluent (yield = 78%, 0.424 g). The tert-
butoxycarbonyl group was deprotected using TFA (1.5 equiv, 0.12 g) in previously mentioned
way and the free amine derivative was obtained (yield = 86%, 0.305 g). On the other hand, N-
carbobenzoxy-L-histidine was synthesized adding benzyl chloroformate (1 eqiv, 0.275 g)
dropwise to NaOH-dioxane solution of L-histidine (1 eqiv, 0.25 g) and stirring the solution
overnight. Afterwards, it was dried and acidified with 1 N HCl and extracted in DCM (yield =
6

Page 7 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
76%, 0.354 g). This (1 equiv, 0.173 g) was conjugated with the previously prepared amine
derivative (1 equiv, 0.305 g) using DCC coupling method. The coupled product was purified in
60-120 mesh silica gel chromatography using 3% methanol in chloroform (v/v) as eluent (yield =
79%, 0.37 g). Next, the benzyl group was deprotected through hydrogenation using Pd on
activated charcoal in THF. Finally, the end product NID (yield = 78%, 0.24 g) was obtained by
filtration and evaporation of the solvent, THF (Scheme S1, ESI).
Preparation of organic nanoparticles. A stock solution (10 mM) of synthesized naphthalimide
derivative was prepared by dissolving the calculated amount of NID in DMSO. The desired
aliquot of this stock solution was added to the varying DMSO-water/aqueous phosphate buffer
(pH = 7.4, 10 mM) solvent systems (0-99% water content) to achieve 10-80 μM of NID, which
was used for all experiments.
Transmission electron microscopy (TEM) study. TEM images were taken in JEOL JEM 2010
microscope. 4 μL solution of NID (10 μM) in 99% water in DMSO (fw = 99 vol%) was drop cast
on a carbon coated 300 mesh copper grid and dried. The negative staining of the copper grid was
done with uranyl acetate solution (1 μL (1% w/v)). TEM images were obtained after keeping the
samples under vacuum for 4 h.
Field-emission scanning electron microscopy (FESEM) study. FESEM images were recorded
in JEOL-6700F microscope. 6 μL solutions of NID (10 μM) and NID + Fe3+ (10 μM NID, 10
M Fe3+) taken in DMSO-water (fw = 99 vol%) was deposited on glass coverslips. The FESEM
images were captured after keeping the samples under vacuum for 3 h.
Atomic force microscopy (AFM) study. Veeco, AP0100 instrument in noncontact mode was
utilized to obtain the AFM images. 10 L of solution of NID (10 M) and NID + Fe3+ (10 μM
7

Journal of Materials Chemistry B
Page 8 of 45
View Article Online
DOI: 10.1039/D0TB02450C
NID, 10 M Fe3+) taken in DMSO-water (fw = 99 vol%) was deposited on mica (1 cm × 1 cm)
and dried overnight.
Fluorescence microscopy study. Fluorescence microscopic images were taken in a IX83
inverted microscope at 40x magnification.10 μL solution of NID (20 μM) in DMSO-water (fw =
99 vol%) was drop cast on glass slides. The solutions were air-dried before imaging.
Dynamic light-scattering (DLS) study. Zen 3690 Zetasizer (Malvern Instrument Ltd.) was used
for measurement of mean hydrodynamic diameter (Dh). The scattering intensity was at 175°
angle and Malvern Zetasizer software was used for data analyzing to determine the Dh (for both
number averaged and intensity averaged distribution) of NID (10 μM) in DMSO-water (fw = 99
vol%) and also the correlogram.
UV-visible study. Solvent dependent UV-vis spectra were recorded on Perkin Elmer Lambda 25
spectrophotometer to understand self-aggregation of NID. UV-visible spectra of NID (50 M) in
DMSO-water solvent mixtures (1 mL) was observed by varying the content of DMSO from
100% to 1% in water. UV-vis spectra of NID (50 μM) were also taken in organic solvents (m-
xylene, o-xylene, toluene, anisole, diethyl ether) with different polarities. Time dependent UV-
vis spectroscopic study was carried out to investigate the stability of NID FONPs. We have
studied the relative absorbance (A/A0) of NID FONPs for 7 days. Here A0 and A were original
absorbance and absorbance at different days, respectively.
1
Solvent-dependent 1H-NMR study. H-NMR spectra of NID (1 mM) in DMSO-d6, 2:1 (v/v)
DMSO-d6-D2O and 1:1 (v/v) DMSO-d6-D2O were recorded on Avance 300 MHz spectrometer
(Bruker).
Photoluminescence study. Fluorescence spectra of NID solutions (excited at (ex) = 350 nm)
were recorded in Varian Cary Eclipse luminescence spectrometer. Emission spectra of NID (10
8

Page 9 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
μM) were monitored in DMSO-water (1 mL) of variable compositions (fw = 0-99 vol %).
Concentration dependent emission spectra were taken in 99% water in DMSO with varying
concentrations (10-90 M) of NID. Emission spectra of NID (50 M) were taken in several
organic solvents (m-xylene, o-xylene, toluene, anisole, diethyl ether) of different polarities.
Photoluminescence study was also performed to know the stability of NID FONPs. We have
studied relative fluorescence intensity (I/I0) of NID FONPs for 7 days. I0 and I represent original
fluorescence intensity and fluorescence intensity at different days. Interaction of aggregated NID
FONPs (50 M) with Fe3+ (FeCl3 solution) was recorded by luminescence study. Stern-Volmer
constant (KSV) of the kinetics of fluorescence quenching of NID FONPs with respect to Fe3+
concentration was found out using the equation: F/F0 = 1 + KSV[Q], where F and F0 are the
fluorescence intensities of NID FONPs in presence/absence of Fe3+ and [Q] denotes
concentration of quencher Fe3+ ion. Fe3+detection limit was evaluated the from the linear curve
obtained from (F0 – F)/F0 vs the Fe3+ concentration and the selectivity of NID FONP (50 M)
against different metal ions having concentration of 500 M was also studied. We carried out
Fe3+ sensing experiments by fluorescence spectroscopy in aqueous buffer solution (phosphate
buffer, 10 mM having pH = 7.4) as well. NID FONPs and Fe3+ stock solutions were prepared in
phosphate buffer medium. This quenching was analyzed quantitatively by using Stern-Volmer
equation like the preceding experiment in Milli-Q water. We recorded the photoluminescence
spectra of the solution of NID FONPs (50 M) at fw = 99 vol% water in DMSO and Fe3+ (500
M) in presence of several Fe3+ chelating agents like 2-aminopyridine, citric acid, L-Dopa,
EDTA, folic acid and glycine having concentration of 500 M and 1000 M. Selectivity
experiments was also carried out taking other monovalent, divalent, trivalent metal ions in
aqueous phosphate buffer (pH = 7.4, 10 mM). Photographs of cuvette containing NID solutions
9

Journal of Materials Chemistry B
Page 10 of 45
View Article Online
DOI: 10.1039/D0TB02450C
(1 mL) of different DMSO-water compositions, NID solutions in presence and absence of
different metal ions in DMSO-water (fw = 99 vol%) and with Fe2+ and H2O2 were taken upon
irradiation under UV lamp (λex = 365 nm).
Quantum yield (QY) measurement. The QY of the unknown fluorophore NID (ϕu) was
measured in comparison to a common fluorophore having a known QY (ϕs) using the following
equation
and
emission intensities of unknown and known samples at same λex are
Refractive indices of solvents are and in which unknown and known sample were
are the absorbance of the unknown and reference sample at respective λex. Integrated
and , respectively.
u
s
dissolved. For the determination of the QY, solutions with similar absorbance (<0.01) were used.
Quinine sulfate in sulfuric acid (0.1 M) was used as the reference having QY (ϕs) = 54.0%.
Time-resolved study. For the time-correlation single photon count (TCSPC) measurement,
samples were excited in NANO-LED IBH-375 and Hamamatsu MCP photomultiplier was used
to collect the fluorescence decays. To perform the experiment, 2 mL solutions of NID (20M)
were prepared in DMSO, DMSO-water (2:3 v/v, 1:99 v/v). All amphiphilic solutions were
excited (λex) at 375 nm and their emission (λmon) was monitored at 390 nm for only DMSO and at
470 nm for other two solvent mixtures. Time-resolved fluorescence decay p(t) was analyzed by
following equation 2
10

Page 11 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
Here, = number of discrete emissive species, = baseline correction (“dc” offset),
exponential factors, and = excited state fluorescence lifetimes for the th component,
respectively. For multiexponential decays, the average lifetime was calculated using equation 3
= pre-
Here,
depicts the contribution of a decay component.
Job’s plot measurements. Two stock solutions of NID and Fe3+ were prepared each having 80
M concentration. Varying volume (0, 125, 250, 375, 500, 625, 750, 875, 1000 L) of NID
solution was mixed with different volume (1000, 875, 750, 625, 500, 375, 250, 125, 0 L) of
Fe3+ solution, respectively having total volume of 1 mL in each vial. Following gentle agitation
of the vials for few minutes, fluorescence spectra were recorded at room temperature.40
Circular dichroism (CD) study. CD spectra of NID (500 μM) in different ratios of DMSO-
water mixtures in absence and presence of Fe3+ (5 mM) at fw = 99% in DMSO and in different
organic solvents like diethyl ether, toluene, anisole, m-xylene and o-xylene were recorded by
using a quartz cuvette of 1 mm path length in a JASCO J-815 CD spectropolarimeter.
Cell culture. B16F10 (cancer cells) and NIH3T3 (non-cancer cells) were brought from NCCS,
Pune. The cells were cultured in DMEM media with 10% FBS, streptomycin and penicillin and
kept in incubator (5% CO2) at 37 °C. Above mentioned cells were used to perform
cytocompatibility and bioimaging experiments. All the cellular experiments were investigated
using NID FONPs at fw = 99 vol%, as they have maximum luminescence compared to other
DMSO-water solution.
MTT assay. Cytotoxicity of NID FONPs in DMSO-water (fw = 99 vol%) was investigated by
MTT assay. In this assay, soluble tetrazolium gets reduced to insoluble formazan by
11

Journal of Materials Chemistry B
Page 12 of 45
View Article Online
DOI: 10.1039/D0TB02450C
mitochondrial dehydrogenase excreted from viable cells. Insoluble formazan was dissolved in
DMSO and estimated spectrophotometrically. The formation of formazan is proportional to the
number of alive cells. Lower the absorbance means the killing of cells. First, NIH3T3 and
B16F10 cells were grown in a 96-well plate (2×104 cells per well) for 24 h. NID FONPs (fw = 99
vol%) having concentration of 10-50 μM was incubated with both the cells for 12 h and 24 h. 10
μL of MTT stock solution (5 mg mL-1) in phosphate buffered saline (PBS) was added and cells
were incubated further for 4 h. The precipitated formazan was dissolved in DMSO and the
absorbance was recorded using Bio Tek1 Elisa Reader at 570 nm. The number of alive cells was
calculated as % viability by following the equation
% viability = (A570 (treated cells) – background)/(A570 (untreated cells) – background) × 100 (4)
Bioimaging. B16F10 and NIH3T3 cells grown in a chamber slide were treated with i) NID
FONPs (25M, fw = 99 vol%), ii) NID FONPs (25M, fw = 99 vol%) + Fe3+(250M), iii) NID
FONPs (25M, fw = 99 vol%) + Fe2+(250 M) for 6 h. Cells were consequently washed with
PBS thrice, and thereafter fixed using paraformaldehyde (4%). The cells were mounted on a slide
using Glycerol (50%), covered with a cover slip and kept for 24 h. Real time bioimaging was
also carried out in chamber slide. Experimental details are available in ESI. Images were taken in
a IX83 inverted microscope at 40x and 20x magnification.
Co-culture experiment: Normal cells (NIH3T3) and cancer cells (B16F10) both were cultured
separately in cell culture flask using DMEM media with 10% FBS, streptomycin and penicillin
and kept in incubator (5% CO2) at 37 °C for 3 days. For co-culture experiment we trypsinized
both type of cells and plated them in each well of 4 well chamber slide as a mixed population
(NIH3T3:B16F10= 1:1) for 24 h in incubation. Then we loaded NID FONPs (25 M, fw = 99
vol%) + Fe2+ (250 M) and incubated it for 6 h. After 6 h incubation cells were consequently
12

Page 13 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
washed with PBS thrice, and thereafter fixed using paraformaldehyde (4%). The cells were
mounted on a slide using glycerol (50%), covered with a coverslip and kept for 24 h. Imaging
was done in a IX83 inverted microscope at 20x magnification.
Flow cytometry. Both the normal cell (NIH3T3) and cancer cell (B16F10) were culturally
grown for 24h. Then we incubated i) NID FONPs (25 M, fw = 99 vol%), ii) NID FONPs (25
M, fw = 99 vol%) + Fe3+ (250 M), iii) NID FONPs (25 M, fw = 99 vol%) + Fe2+ (250 M)
with both type of cells for 6h. After 6h, we washed both the incubated cells with DMEM media
and PBS buffer to remove the excess compounds from cell medium. We used trypsin to detached
the cells from culture flask and centrifuge it for 5 min. After centrifugation we took the
precipitated cells suspended in PBS (500 μL) and carried out the cytometric experiment. BD
FACS AriaTM III flow cytometer was employed to inspect the cells at emission wavelength using
a 510 ± 40 nm bandpass filter upon excitation at 405 nm.
Results and discussion
Molecular designing of NID. Aggregation-induced emission (AIE) is a direct result of restricted
intramolecular rotation of luminogens as clusters/aggregates. Naphthalimide core is a widely
known fluorophore unit considering its unique photophysical properties and high condensed-
state emission behavior that is being explored extensively in various applications.41 To date,
naphthalimide derivatives have been reported with high quantum yields in solid or aggregated
state in a large number contrary to aggregation caused quenching (ACQ).42 N-functionalization
of naphthalimide or inclusion of different organic moieties at fourth position of NI core is
reported to influence AIE owing to the restricted intramolecular rotation (RIR) and loss in the
planarity of molecule.42-45 As the naphthalimide core is electron acceptor in nature, it can be
functionalized with π-conjugated electron-donating units to have intramolecular charge transfer
13

Journal of Materials Chemistry B
Page 14 of 45
View Article Online
DOI: 10.1039/D0TB02450C
from donor ligands to electron deficient molecular scaffold, which may have a significant
contribution towards achieving AIE property.42 Considering these parameters, we have designed
naphthalimide-based amino acid-linked amphiphilic molecule (NID, Fig. 1) where the
hydrophobic aromatic ring of the naphthalimide core may assist during self-aggregation.
Chirality of the L-aspartic acid and histidine is expected to induce RIR in the self-aggregated
state owing to their twisted nature in the molecular backbone.15,19 Consequently, the molecule
will be highly emissive as it is expected to relax through radiative pathways in the condensed
state. To explore the AIE property of the luminogen in selective metal sensing, histidine was
incorporated in the NID structure via amide linkage having imidazole 'N' and amino 'N' free to
serve as a metal chelating moiety. Its co-ordination with a metal ion could cause an enhancement
of the fluorescence emission, called chelation enhanced fluorescence effect (CHEF) or a
quenching of the fluorescence, called chelation enhancement quenching effect (CHEQ).46
13
Characterization data of synthesized NID (1H, C NMR and MALDI-TOF mass spectra) were
given in ESI (Fig. S1-S3).
Microscopic investigations and DLS study of the self-assemblies of NID. The synthesized
NID was soluble in DMSO. The transparent solution turned to translucent in presence of 60%
and above water content (Fig. 2). The formation of translucent solution presumably indicates the
initiation of self-assembly by NID amphiphiles. With increase in the water proportion up to 99%,
translucency of the NID solution got significantly enhanced probably owing to the development
of self-aggregates. Several microscopic techniques were employed to examine the structures of
the possible self-assembled units formed by NID derivative. TEM images of NID (10 μM, fw =
99 vol%) showed spherical aggregates with diameter of 70-100 nm (Fig. 3a). The formation of
organic nanoparticles by NID (10 μM, fw = 99 vol%) with spherical morphology having
14

Page 15 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
dimension of 90-120 nm was also confirmed from FESEM images (Fig. 3b). The observed
diameter of the spherical aggregates in the corresponding AFM image was also in a range of 70-
120 nm (Fig. 3c). Fluorescent microscopic images exhibited green emitting dots (Fig. 3d).
Additionally, mean hydrodynamic diameter (Dh) of NID organic nanoparticles was found to be
in the range of 50-250 nm with an average diameter of ∼100 nm in dynamic light-scattering
(DLS) experiment for both number averaged and intensity averaged distribution (Fig. 3e, Fig.
S4a, ESI). The correlogram with good Cumulant Fit for NID is provided in Fig. S4b, ESI. All the
microscopic evidences and the DLS data were mostly in concurrence with each other.
Spectroscopic study for self-assembly of NID amphiphile. Self-aggregation of NID towards
the formation of organic nanoparticles was initially examined by UV-visible spectroscopy in
varying compositions of DMSO-water. In pure DMSO (molecularly dissolved state), NID (50
μM) showed a characteristic UV band in the range 330-360 nm having λmax at 343 nm, which
may be ascribed to the π-π* transition of NID chromophore (Fig. 4a).47 With gradual increase in
the water amount within DMSO, the absorption maxima slightly but steadily got red-shifted. At
fw = 99 vol%, the λmax was observed at 347 nm (Fig. 4a). This red shift of λmax implied the
generation of self-aggregates possibly via J-aggregation (head-to-tail fashion arrangement). The
extinction coefficients of NID in DMSO, 20 vol%, 40 vol%, 60 vol%, 80 vol%, 99 vol% water
in DMSO were 0.009, 0.0096, 0.01, 0.0104, 0.0114, 0.0122 M-1 cm-1, respectively.
Involvement of different non-covalent interactions during self-assembly (e.g.
hydrophobic interaction and π-π stacking) was investigated by solvent-dependent 1H-NMR study
(Fig. 4b). The aromatic protons of NID exhibited sharp NMR signal in the region δ = 8.36-8.55
ppm (aromatic core of NID) and δ = 6.74-8.09 ppm (naphthyl ring) in DMSO-d6, in its non-
assembled state (Fig. 4b). With increasing the D2O content gradually from 0% to 30% and 50%,
15

Journal of Materials Chemistry B
Page 16 of 45
View Article Online
DOI: 10.1039/D0TB02450C
NMR signals of both the NID aromatic core and the naphthyl group got broadened and shifted to
δ = 7.86-8.29 ppm and δ = 6.24-7.81 ppm, respectively (Fig. 4b). The π-π stacking between the
chromophore and the naphthyl moieties of the molecules is facilitated with the addition of D2O
that eventually results in the upfield shift in the 1H-NMR signals indicating the formation of self-
aggregated structures.
Investigation of the emissive nature of organic nanoparticles. Intrigued by the change in the
UV-vis spectral nature of NID during self-aggregation, we investigated the emission behavior of
NID in DMSO-water solvent mixtures. Emission spectrum of NID (10 μM) in DMSO showed
almost negligible fluorescence intensity (Fig. 5a). Moreover, no significant fluorescence
intensity was observed with increasing water content up to 50% within the DMSO solution (fw =
50 vol%). Interestingly, an excimer peak appeared having emission maxima at 470 nm at 60%
water content (fw = 60 vol%). Further addition of water resulted in stronger emission for self-
aggregated NID having its highest intensity at fw = 99 vol% (Fig. 5a). This is in concurrence with
the absorption spectroscopic behaviour of NID (Fig. 4a). NID Amphiphiles started engaging in
self-assembly apparently through the stacking of the chromophoric units with increase in water
content. It resulted in excimer peak formation at 470 nm owing to AIE. The shift in the emission
maxima is a result of changes in electronic properties of the excited-states that take place in
proceeding from the solution to the aggregated phase which can be classified as excimer
formation in the self-assembled state of -conjugated systems arising due to - interactions
between the respective naphthalimide chromophores.17,19 The emission intensity of NID was
many-fold higher at fw= 99 vol%, than at pure DMSO (Fig. 5a,b). Relative fluorescence quantum
yield of NID at fw= 99 vol% is calculated to be 8% (with respect to quinine sulfate hydrate). This
AIE behavior was also examined by irradiating NID in different DMSO-water compositions (1
16

Page 17 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
mL) under UV lamp having λex = 365 nm (Fig. 2,5b). DMSO solution of NID (50 M) showed
no significant fluorescence and from 60% water content a bluish green emission was witnessed,
which became maximum at fw= 99 vol%, due to AIE via the excimer formation (470 nm). As the
emission intensity was maximum at fw= 99 vol% in DMSO-water binary solvent system, all
further experiments were carried out using NID FONPs prepared in 1:99 DMSO-water solution.
Subsequently, concentration-dependent fluorescence spectra of NID at fw= 99 vol% were
recorded within a concentration range of 10-90 μM (Fig. 5c). With increase in NID
concentration, the luminescence intensity of the characteristic excimer band (λem at 470 nm)
consistently enhanced and became saturated at and above 70 μM (Fig. 5c). This observation
delineates that with increase in the concentration of NID, enhancement in the AIE was noted
possibly due to the self-aggregation of a greater number of amphiphilic molecules which became
saturated beyond a specific concentration.
For better understanding of the AIE phenomenon in NID, TCSPC study was carried out
with picosecond excitation at 375 nm for NID (20 μM) in pure DMSO, 60% and 99% water
content in DMSO (Fig. 5d). In all cases, triexponentially fitted decay curves were attained. For
DMSO solution of NID, the emission was monitored at λmon = 390 nm and the average lifetime
value (⟨τ⟩) were 0.0456 ns (Table 1). The ⟨τ⟩ value monitored at λmon = 470 nm for fw = 60 vol %
and 99 vol% in DMSO, was found to be 4.04 ns and 6.87 ns, respectively (Table 1). Pre-
associated excimer in the AIE resulted in the long decay lifetime at higher percentage of water.
Furthermore, we also analyzed the influence of intramolecular charge transfer process
(ICT) on the AIE phenomenon by checking the UV-vis and emission spectra of NID (50 μM) in
varying organic solvents having distinct polarities (Fig. 6a-c). Alteration in the solvent from non-
polar xylene to polar diethyl ether did not show any considerable changes in the absorbance
17

Journal of Materials Chemistry B
Page 18 of 45
View Article Online
DOI: 10.1039/D0TB02450C
maxima (Table 2 and Fig. 6a). The molar extinction coefficients of NID in diethyl ether, toluene,
anisole, m-xylene and o-xylene were 0.0132, 0.013, 0.011, 0.0056, 0.0034 M-1 cm-1,
respectively. Nevertheless, the λem significantly red-shifted with gradual enhancement in polarity
of solvent (Table 2 and Fig. 6b). The Stokes shift (Δν) of NID in different solvents was
determined and the solvatochromic property was also quantitatively expressed by the Lippert-
Mataga equation.48-49
where
Planck’s constant, = speed of light, = Onsager solvent cavity,
moment and = ground-state dipole moment. The solvent polarity parameter (orientational
= maximum absorbance wavenumber,
= maximum emission wavenumber,
=
= excited-state dipole
polarizability) , represented in eq , is the measure of solvent polarity.
where = static dielectric constant and = optical refractive index of the solvent. For NID
linearity between Δν vs Δf with positive solvatochroism was observed, which represented ICT
from the naphthyl unit (donor part) to the NID core (acceptor part) (Fig. 6c, Scheme 1).
We also performed time dependent UV-vis spectroscopic study as well as
photoluminescence study to know the stability of NID FONPs. The relative absorbance (A/A0)
and relative fluorescence intensity (I/I0) of NID FONPs were recorded with time (Fig. S5a-b,
ESI). Here A0 and A were original absorbance and absorbance at different days, respectively.
Similarly, I0 and I represent original fluorescence intensity and fluorescence intensity at different
days. No significant change in the relative absorbance and fluorescence intensities was noted for
18

Page 19 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
7 days. Also no precipitation was observed in the translucent NID solutions for 7 days (Fig. S5c,
ESI). Hence, these FONPs were stable for a considerable time period.
Fluorimetric sensing of Fe3+. After ensuring the formation of NID FONPs, we intended to
explore the emissive behavior of the FONPs in Fe3+ sensing owing to the presence of histidine in
the amphiphile backbone. Histidine is well known chelating agent for Fe3+.50 Among all the
transition metals, iron plays vital roles in oxygen uptake, electron transfer, cellular metabolism,
enzyme catalysis, and many other biochemical processes. Therefore, we aim to utilize NID
FONPs towards selective detection of Fe3+. Consequently, the fluorescence spectra of NID
FONPs (50 M) was recorded in DMSO-water, fw= 99 vol% (1 mL) upon titration with Fe3+
(aqueous solution of FeCl3) (Fig. 7a). The emission intensity of NID FONPs at ex = 470 nm was
found to be decreased gradually along with the increase in concentration of Fe3+ from 10 M
onwards. The quenching effect by Fe3+ steadily increased up to 560 M and thereafter no
significant changes in quenching effect was noted (experimentally investigated up to 650 M).
This quenching by Fe3+ on the emission of NID FONPs was noted possibly due to the complex
formation between Fe3+ and the histidine residue of the NID amphiphile (Scheme 1). In a wide
variety of metal complexes, the formally forbidden intersystem crossing (ISC) becomes faster
due to the presence of a paramagnetic metal ion in the proximity of the fluorophore. In the
present study, the observed quenching of fluorescence by Fe3+ ions may be ascribed due to the
cooperative influence of size compatibility as well as the paramagnetic nature of Fe3+ which
result in chelation-enhanced quenching (CHEQ).46 The coordination between Fe3+ and histidine
took place among one of the N atoms from imidazole ring and another N atom from the free
amine.50 The quenching of NID FONPs by Fe3+ was analyzed quantitatively by using Stern-
Volmer equation,
19

Journal of Materials Chemistry B
Page 20 of 45
View Article Online
DOI: 10.1039/D0TB02450C
F/F0 = 1 + KSV[Q]
(7)
In this equation, F and F0 are the fluorescence intensities of NID FONP in the presence/absence
of Fe3+ respectively; KSV is the Stern-Volmer constant; and [Q] is quencher concentration. The
change in F/F0 of NID FONP with Fe3+ up to 70 M is observed and the Stern-Volmer plot
showed linearity in fluorescence quenching of NID by Fe3+ with KSV of 0.0035 M-1 (Fig. 7b).
As the NID FONP displayed a sensitive turn off response to Fe3+, thereby we measured
the detection limits of Fe3+. The emission intensity of NID FONPs at 470 nm decreased in
proportion to the concentration of Fe3+ ranging from 10 to 560 M (Fig. 7a). The sensitivity was
measured based on the relationships between the ratio of fluorescence emission intensity at 470
nm in presence and absence of Fe3+. A good linear behavior was obtained between (F0 – F)/F0
and Fe3+ in the lower concentration range (up to 70 M) where F0 and F were emission
intensities of NID FONPs at 470 nm in the absence and presence of Fe3+, respectively (Fig. 7c).
The linear regression equation was F0 - F/F0 = -8.78645 + 0.00351x concentration of Fe3+ (M);
correlation coefficient (R2) of 0.99114. The limit of detection (LOD) of Fe3+ was found to be
12.5±1.2 M as calculated based on 3σ/S where σ was the standard deviation and S was the
slope of the calibration curve (experiment was repeated thrice).
To determine the binding ratio between luminescent NID and Fe3+, the stoichiometric
ratio of NID and Fe3+ was calculated through Job's plot (Fig. 7d and Fig. S6, ESI).40 The
stoichiometric ratio of NID and Fe3+ was established by plotting the mole fraction (x) of NID
and the product of emission intensity change (F0 - F, where F0 and F were emission intensities of
NID FONPs at 470 nm in the absence and presence of Fe3+, respectively) and the mole fraction
(x) of NID. Fluorescent intensities of 1 mL solutions of NID + Fe3+ in different ratios, each
having total concentration of 80 M was measured. The product of F0 - F and mole fraction (x)
20

Page 21 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
of NID went through maxima at x = 0.501 as shown in Fig. 7d, indicating the formation of 1:1
stoichiometric complexes between NID and Fe3+ that led to the quenching of the emission
intensity of NID FONPs.
After confirming the Fe3+ sensing by NID FONP, we were curious to investigate the
selectivity of NID against the other metal ions. This is a very essential parameter to estimate the
performance of a new fluorescence sensor. To this end, 1 mL of NID FONP solution (50 M)
was treated with different metal ions. Emission spectra of NID were recorded in the presence of
different monovalent (Na+, K+), divalent (Co2+, Zn2+, Fe2+, Pb2+, Ni2+, Mn2+), trivalent (Cr3+, V3+)
metal ions and Fe3+, having concentration of 500 M for each metal ions. Interestingly, none of
these metal ions could quench the fluorescence of NID FONP significantly except Fe3+ (Fig. S7,
ESI). This experiment was repeated thrice and the error bar reflects the standard deviation. The
effect of the different metal ions on the emission of the NID FONPs was also examined by
irradiating the samples (NID + metal ion solution in 99% water in DMSO) under UV lamp (λex =
365 nm) (Fig. S8, ESI). The bluish green emission of NID FONPs remains unchanged for all the
used metal ions except Fe3+ containing solution. In presence of Fe3+, the emission intensity got
drastically quenched indicating further the selective sensing of Fe3+ by NID FONPs.
All these preceding experiments were carried out using Milli-Q water. To investigate the
complexation of Fe3+ with NID has any influence on the pH of the medium and subsequently the
AIE property of NID FONPs, we carried out Fe3+ detection in aqueous phosphate buffer solution
(pH = 7.4, 10 mM). NID FONPs and Fe3+ stock solutions were prepared in phosphate buffer
medium. With gradual increase in Fe3+ concentration, the fluorescence intensity got gradually
quenched up to 560 M in concurrence with the preceding observation (Fig. S9a, ESI). No
further quenching was observed up to 650 M of Fe3+. This quenching was similarly analyzed
21

Journal of Materials Chemistry B
Page 22 of 45
View Article Online
DOI: 10.1039/D0TB02450C
quantitatively by using Stern-Volmer equation (7). The Stern-Volmer plot of change in F/F0 of
NID FONP with Fe3+, showed linearity in fluorescence quenching of NID by Fe3+ up to 55 M
with KSV of 0.0037 M-1, which was 0.0035 M-1 in case of Milli-Q water. We also carried out
the selectivity experiments taking other monovalent, divalent, trivalent metal ions in aqueous
buffer medium (pH = 7.4, 10 mM). Here also, NID FONPs showed similar selectivity towards
Fe3+ (Fig. S9c,d, ESI) against other tested metal ions.
Next, we wanted to investigate the chelation between NID FONPs and Fe3+ with respect
to other Fe3+ binding agents. To this end, we recorded the photoluminescence spectra of Fe3+
(500 M) containing solution of NID FONPs (50 M) at fw= 99 vol% water in DMSO treated
with several Fe3+ chelating agents like 2-aminopyridine, citric acid, L-Dopa, EDTA, folic acid
and glycine having concentration of 500 M and 1000 M, respectively. As mentioned earlier,
the fluorescence of NID FONPs (50 M) gets quenched in presence of Fe3+ (500 M) (Fig. 7a).
However, no significant increase or recovery in the fluorescence intensity took place in presence
of 2-aminopyridine, citric acid, L-Dopa, EDTA, folic acid and glycine (Fig. S10, ESI). Hence, it
may be inferred that the co-ordination between NID FONPs and Fe3+ is quite strong. We again
carried out the experiment with same concentration of NID FONPs (50 M) and Fe3+ (50 M)
with twice the concentration of EDTA (100 M) both in aqueous and phosphate buffer (pH=7.4,
10 mM) medium (Fig. S11, ESI). Concentration has been kept equivalent to eliminate the
fluorescence suppressing phenomena. No regain of fluorescence took place in presence of EDTA
to the NID FONP solution treated with Fe3+. This further confirmed the strong binding nature of
NID FONPs to Fe3+.
As mentioned before, oxidative stress is an important parameter for a living cell. The
oxidative stress can be varied owing to the uncontrolled production of ROS through Fe2+/Fe3+
22

Page 23 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
transition in presence of hydrogen peroxide.32-33 In view of the unique reactivity of iron at
different redox states, it would be highly important to detect Fe2+/Fe3+ transition. Initially, we
investigated this Fe2+/Fe3+ transition in a cuvette using NID FONP in presence of H2O2 (as an
oxidizing agent). In a cuvette containing 50 M of NID FONP at fw= 99 vol% (1 mL), requisite
volume of Fe2+ (FeCl2) solution was added so that resultant Fe2+ concentration became 500 M.
The NID FONP solution was irradiated under UV lamp at λex = 365 nm before and after addition
of Fe2+ and no change in the bluish green emission was noted (Fig. S12, ESI). Interestingly, the
addition of equivalent amount of H2O2 (500 M) resulted in the complete quenching of the
emission (Fig. S12, ESI) possibly by Fe3+ that formed through oxidation of Fe2+ in presence of
H2O2. Notably, addition of H2O2 (500 M) to the only NID FONPs solution (in absence of Fe2+)
did not exhibit any marked change in its emission.
Influence of Fe3+ on the morphology of self-aggregated NID FONPs. Selective and sensitive
detection of Fe3+ by NID FONPs made us curious to investigate whether coordination of Fe3+
with histidine moiety of NID has any influence on the morphology of self-aggregates. FESEM
images taken for 10 M solution of NID +Fe3+ (mixed in equimolar proportion) at fw= 99 vol%
showed spindle shaped morphologies having length of  200 nm and width of 60-70 nm (Fig.
8a). Similar morphologies were observed in the AFM images where the aggregates had length in
the range of 200-250 nm and width of 60-80 nm (Fig. 8b). It is evident that chelation of NID
with Fe3+ led to the morphological transformation from spherical to spindle in the structure of
self-aggregates.
Circular dichroism (CD) study. The effect of chiral amino acid moieties in the self-assembly of
NID was investigated by CD spectroscopy. In pure DMSO (molecularly dissolved state) NID
showed chirality without any characteristic CD signal (Fig. S13a, ESI). Upon gradual increase in
23

Journal of Materials Chemistry B
Page 24 of 45
View Article Online
DOI: 10.1039/D0TB02450C
the water content, a change in the CD spectral pattern with the possible emergence of chiral bias
was noted. At the highest percentage of water (fw = 99%) in DMSO, the CD spectra with a
positive band at 240 nm and a negative band at 248 nm was observed (Fig. S13a, ESI) which
resembles the possible -sheet type (not by the position of the peaks) arrangement of the self-
aggregates. This manifestation of chiral bias possibly resulted from the supramolecular
organization of the amino acid containing chiral building block, NID. The positive band
observed in the CD could be attributed to the -* transition of the amide bond and negative
band at longer wavelength to n-* transition of the same. With introduction of Fe3+ (5 mM), this
characteristic peak disappeared (Fig. S13b, ESI) as a result of the alteration in the
supramolecular arrangement of the self-aggregates that led to the change in morphology of the
self-aggregates (Fig. 8).
Cytocompatibility of NID derivative-based FONPs. The cytocompatibility of NID FONPs
against mammalian cells is crucial for exploring the AIE-gens in bioimaging and Fe3+ sensing
within cellular environment. The cytocompatibility of NID FONPs (10-50 μM) prepared at fw =
99 vol% was examined by MTT assay in non-cancer (NIH3T3) and cancer (B16F10) cells. In
each case, after 12 h and 24 h incubation 85-90% and ~82-90% eukaryotic cells (normal and
cancer cells) were found to be alive (Fig. S14, ESI). Therefore, these NID FONPs were found to
be adequately biocompatible and suitable for cellular imaging.
Bioimaging. NID-derived biocompatible FONPs were utilized in bioimaging of the mammalian
cells. Both non-cancerous (NIH3T3) and cancerous (B16F10) cells were incubated with 25 μM
of NID FONPs (fw= 99 vol%) for 6 h. Fluorescence microscopic images showed green
fluorescence inside both the cells indicating significant internalization of the NID FONPs within
the cells and mean fluorescence intensity was found to be 1067 for NIH3T3 and 1035 for
24

Page 25 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
B16F10 (Fig. 9a-c, 10a-c). This cell-staining ability of the FONPs in the visible range has
potential utility in diagnostics as bio-probes. Fluorescence microscopic images were further
taken to investigate the ability of NID FONPs towards sensing Fe3+ as well as monitoring
Fe2+/Fe3+ transition inside cells. Incubation of exogenous Fe3+ (250 μM) along with NID FONP
with NIH3T3 and B16F10 cells exhibited substantially quenched fluorescence inside both the
cells having mean fluorescence intensity of 133 and of 204, respectively (Fig. 9d-f, 10d-f,
Scheme 1). Moreover, as discussed above, Fe2+ ions are highly relevant in case of oxidative
stress where through Fenton reaction Fe2+ ion reacts with hydrogen peroxide to generate highly
reactive oxygen species and got oxidized to Fe3+ state. In this context, cancer cells are known to
possess more H2O2 compared to normal alive cells.51-53 Considering this fact, we incubated the
mentioned mammalian cells with Fe2+ (250 μM) and NID FONP for 6 h. In case of non-
cancerous NIH3T3 cells, no significant quenching in the green emission of NID FONP was
observed. The mean fluorescence intensity was found to be 1010 (Fig. 9g-i), which is
comparable to that of native NID FONPs. Notably, in B16F10 cancer cells, green fluorescence
of NID FONP within the cells got significantly quenched with a low mean fluorescence intensity
of 342. This was possibly due to the oxidation of included Fe2+ to Fe3+ by H2O2 inside cancer
cells (Fig. 10g-i), which could not take place in the non-cancerous one owing to the less H2O2
content. We also carried out real time bioimaging of NIH3T3 and B16F10 cells taken in chamber
slides by incubating NID FONPs (25 M, fw = 99 vol%) for 3 h and 6 h and again further in
presence of Fe2+ (250 M) and Fe3+ (250M) for 6 h (Fig. S15 and S16, ESI). We observed
similar type of results in concurrence to fixed cells as mentioned above. To demonstrate it
further, we co-cultured both the normal cell (NIH3T3) and cancer (B16F10) cells together and
incubated the cells with Fe2+ (250 μM) and NID FONP (25 μM) for 6 h. In contrast to the
25

Journal of Materials Chemistry B
Page 26 of 45
View Article Online
DOI: 10.1039/D0TB02450C
number of cells present in the bright field image, comparatively lower number of cells exhibited
green luminescence under fluorescence microscopic image in the mixed population of both non-
cancer (NIH3T3) and cancer (B16F10) cells (Fig. S17, ESI). Fe2+ upon internalization within
cancer cells easily got oxidized to Fe3+ state through Fenton reaction because of the high H2O2
content. Consequently, it quenched the fluorescence of included NID FONPs inside the cancer
cells (B16F10). However, this same Fe2+/Fe3+ transition and subsequently fluorescence
quenching of NID FONPs did not take place in case of non-cancer cells (NIH3T3) owing to its
lack of H2O2 content. This result corroborates with the fact that in cancer cells the fluorescence
of NID FONPs quenched in presence of Fe3+. Hence, this newly developed NID FONP can be
successfully utilized in selective sensing Fe3+ inside and outside of cellular environment as well
as for monitoring Fe2+/Fe3+ transition selectively inside cancer cells as a marker of oxidative
stress.
Conclusion
In this present work, we have synthesized naphthalimide based L-histidine appended amphiphile
(NID) which formed bluish green emitting FONPs in 99% water-DMSO binary solvent system
via J-type aggregation. The amphiphile showed negligible emission in DMSO (non self-
assembled state) but became highly emissive at 60% and above water content at the aggregated
state. Aggregation-induced emission was observed through excimer formation, having the
emission maximum at 470 nm upon excitation at 350 nm. Among the various tested metal ions,
presence of Fe3+ led to the selective quenching of the emission intensity of NID FONPs due to
1:1 stoichiometric complex formation with the histidine residue. Consequently, the spherical
morphology of the self-aggregated organic particle transformed to spindle shaped particles as
confirmed by microscopic techniques. NID FONP was used as an efficient AIE based turn-off
26

Page 27 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
sensor for Fe3+ with a LOD of 12.5±1.2 M having high selectivity over other metal ions.
Concurrently, NID FONP was successfully employed for bioimaging of Fe3+ ions via
fluorescence quenching within living cells as well as detecting Fe2+/Fe3+ transition selectively
inside cancer cell due to its high H2O2 content. Hence, NID FONP can be a selective diagnostic
probe for cancer cells.
Electronic supplementary information (ESI) available.
1
Synthetic scheme of NID, comparison of Fe3+ sensors, characterization data of NID, H-
NMR,13C-NMR, mass spectrum of NID, intensity averaged distribution and correlogram of NID
FONPs obtained by at fw = 99 vol%, stability of NID FONPs, emission spectra of different mole
fraction of NID/ Fe3+at fw = 99 vol%, emission spectra and photographs of UV irradiated NID
FONPs in presence of different metal ions, Fe3+ sensing and selectivity studies in phosphate
buffer medium, emission spectra of NID/ Fe3+ at fw = 99 vol% in presence of 2-aminopyridine,
citric acid, L-Dopa, EDTA, folic acid and glycine, emission spectra of of equimolar NID/ Fe3+ at
fw = 99 vol% in presence of EDTA, photographs of UV irradiated NID FONPs in presence and
absence of Fe2+ and H2O2, CD spectra of NID in different ratios of DMSO-water and at fw = 99%
in DMSO in presence of Fe3+, MTT-based % cell viability of NID FONPs in NIH3T3 and
B16F10 cells, real time imaging NID FONPs included NIH3T3 and B16F10 cells, bright-field
and fluorescence microscopic image of co-cultured NIH3T3 and B16F10 cells after 6 h
incubation with NID FONPs + Fe2+.
Acknowledgements
P.K.D. is thankful to Science and Engineering Research Board (SERB), Department of Science
and Technology (DST), India for financial assistance (No. EMR/2017/000656). D.S. and M.C.
acknowledge CSIR, India and UGC, India, respectively for their Research Fellowships.
27

Journal of Materials Chemistry B
Page 28 of 45
View Article Online
DOI: 10.1039/D0TB02450C
References
(1) J. Wu, W. Liu, J. Ge, H. Zhang and P. Wang, Chem. Soc. Rev., 2011, 40, 3483-3495.
(2) H. Kobayashi, M. Ogawa, R. Alford, P. L. Choyke and Y. Urano, Chem. Rev., 2010, 110,
2620-2640.
(3) B. Kong, A. Zhu, C. Ding, X. Zhao, B. Li, and Y. Tian, Adv. Mater., 2012, 24, 5844-5848.
(4) X. Li, X. Gao, W. Shi and H. Ma, Chem. Rev., 2014, 114, 590-659.
(5) Z. Liu, W. He and Z. Guo, Chem. Soc. Rev., 2013, 42, 1568-1600.
(6) J. P. Nandre, S. R. Patil, S. K. Sahoo, C. P. Pradeep, A. Churakov, F. Yu, L. Chen, C.
Redshaw, A. A. Patil and U. D. Patil, Dalton Trans., 2017, 46, 14201-14209
(7) Y. Wang, M. Gao, Q. Chen, F. Yu, G. Jiang and L. Chen, Anal. Chem., 2018, 90, 16, 9769-
9778.
(8) J. Malkin, CRC: Boca Raton, FL, 1992.
(9) J. Luo, Z. Xie, J. W. Y. Lam, L. Cheng, H. Chen, C. Qiu, H. S. Kwok, X. Zhan, Y. Liu, D.
Zhu, and B. Z. Tang, Chem. Commun., 2001, 1740-1741.
(10) Y. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Soc. Rev., 2011, 40, 5361-5388.
(11) J. Mei, N. L. C. Leung, R. T. K. Kwok, J. W. Y. Lam and B. Z. Tang, Chem. Rev., 2015,
115, 11718-11940.
(12) J. Qian and B. Z. Tang, Chem, 2017, 3, 56-91.
(13) M. Gao and B. Z. Tang, ACS Sens., 2017, 2, 1382-1399.
28

Page 29 of 45
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/D0TB02450C
(14) D. D. La, S. V. Bhosale, L. A. Jones and S. V. Bhosale, ACS Appl. Mater. Interfaces, 2018,
10, 12189-12216.
(15) A. K. Ghosh, P. Choudhury and P. K. Das, Langmuir, 2019, 35, 15180-15191.
(16) B. K. An, S. K. Kwon, S. D. Jung and S. Y. Park, J. Am. Chem. Soc., 2002, 124, 14410-
14415.
(17) M. Kumar, and S. J. George, Nanoscale, 2011, 3, 2130-2133.
(18) A. Kaeser, I. Fischer, R. Abbel, P. Besenius, D. Dasgupta, M. A. J. Gillisen, G. Portale, A.
L. Stevens, L. M. Herz and A. P. H. J. Schenning, ACS Nano, 2013, 7, 408-416.
(19) P. Choudhury, S. Sarkar and P. K. Das, Langmuir, 2018, 34, 14328-14341.
(20) X. Zhang, S. Wang, L. Xu, L. Feng, Y. Ji, L. Tao, S. Li and Y. Wei, Nanoscale, 2012, 4,
5581-5584.
(21) M. Ahmed, M. Faisal, A. Ihsan and M. M. Naseer, Analyst, 2019, 144, 2480-2497.
(22) Z. Wang, T. Y. Yong, J. Wan, Z. H. Li, H. Zhao, Y. Zhao, L. Gan, X. L. Yang, H. B. Xu
and C. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 3420-3425.
(23) S. Dai, C. Schwendtmayer, P. Schurmann, S. Ramaswamy and H. Eklund, Science, 2000,
287, 655-658.
(24) A. Atkinson and D. R. Winge, Chem. Rev., 2009, 109, 4708-4721.
(25) E. C. Theil and D. J. Goss, Chem. Rev., 2009, 109, 4568-4579.
(26) D. A. Weinstein, C. N. Roy, M. D. Fleming, M. F. Loda, J. I. Wolfsdorf and N. C. Andrews,
Blood, 2002, 100, 3776-3781.
29