A lysosome-specific near-infrared fluorescent probe for: In vitro cancer cell detection and non-invasive in vivo imaging

Article abstract of DOI:10.1039/c9cc07322a

Near-infrared (NIR) fluorescent probes have been developed as potential bio-materials having profound applications in diagnosis and clinical practice. Herein, we wish to disclose a highly photostable ultra-bright NIR probe for the specific detection of lysosomes in numerous cell lines. Furthermore, the applicability of the developed NIR probe was evaluated for in vivo imaging.

Full text of DOI:10.1039/c9cc07322a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A lysosome-specific near-infrared fluorescent
probe for in vitro cancer cell detection
and non-invasive in vivo imaging†
Cite this: Chem. Commun., 2019,
55, 14182
Received 18th September 2019,
Accepted 27th October 2019
ab
abc
Rakesh Mengji,
Chiranjit Acharya,a Venugopal Vangalaab and Avijit Jana
*
DOI: 10.1039/c9cc07322a
rsc.li/chemcomm
Near-infrared (NIR) fluorescent probes have been developed as (PET), X-ray radiography, computed tomography (CT) and ultra-
potential bio-materials having profound applications in diagnosis sound imaging as diagnostic tools for cancer detection. In spite of
and clinical practice. Herein, we wish to disclose a highly photo- the certain advantages of these techniques, which include user-
stable ultra-bright NIR probe for the specific detection of lyso- friendliness, easy availability and long-range tissue penetration,
somes in numerous cell lines. Furthermore, the applicability of the they suffered from foremost limitations, which include lack of
developed NIR probe was evaluated for in vivo imaging.
selectivity and sensitivity, inadequate efficiency for envisioning real-
time dynamics and radiation hazards.12–14 Therefore, it is worth
The fluorescence imaging technique to visualize living cell demanding to develop a quantized technique having limited or no
processes or even a whole organism in real-time has been revealed cellular invasion with high selectivity and sensitivity along with
to be the central research interest in recent years.1–3 However, the excellent resolution modality for the detection of cancer.13 As a
major shortcomings of the fluorescence imaging technique are low result, NIR fluorescent imaging has been introduced as an
tissue penetration depth, nonspecific background fluorescence and emerging imperative for the detection of cancer at an early
inadequate photostability of the fluorophore. Among various stage.6,14 Hence, there is an urgent need for the development of
fluorescence imaging techniques, near-infrared (NIR) fluorescence highly stable NIR fluorophores with high specificity to cancer
imaging has been found to be prudential due to its excellent deep- analytes for analytical and clinical optical imaging.9
tissue penetration potential and deep-tissue imaging without any
Several NIR fluorescent imaging probes comprising both organic
invasion.4 Therefore, a NIR fluorophore with a high specificity to a and inorganic fluorophores have been illustrated in literature in
specific bio-analyte is highly desirable to visualize the real time recent times. Inorganic NIR probes, including quantum dots and
dynamics of an associated disease.5,6 Recently, several NIR probes other inorganic nanoparticles (NPs), were found to be expensive
have been reported for various applications, including cancer with considerable cytotoxicity, which triggered a switch in research
diagnosis.1,7–9 However, the accurate detection of cancer analytes focus to the search for suitable organic NIR probes.15–20
still remains challenging owing to the lack of suitably functiona-
Among a number of organic NIR fluorescent probes, perylene
lized NIR probes. In recent years, the epidemiology of cancer has derivatives have attained considerable interest since they meet the
been found to be one of the major causes of reduced mortality criteria for good NIR probes, which include red to NIR excitation,
index.10 Prudential cancer therapy involves early detection followed high fluorescence quantum yields, high absorption coefficients,
by parallel treatment of cancer.10,11 However, it needs suitable large Stokes shifts, high stability in biological systems, exceptional
diagnostic development, which is a demanding task in modern stability against photo-bleaching and synthetic simplicity.6,21–23
cancer research.11 The advancement of science and technology With this idea, several attempts have been made to synthesize a
has offered many sophisticated imaging techniques, including perylene-based fluorophore, but only a few have been actualized for
magnetic resonance imaging (MRI), positron emission tomography in vivo deep tissue bio-imaging.5,6,21–23 Herein, we wish to disclose a
perylene monoimide (PeIm)-based new NIR fluorescent probe,
a Department of Applied Biology, CSIR-Indian Institute of Chemical Technology,
which could selectively and non-invasively detect and sense cancer
cells over noncancerous cells in vitro and could also be a practical
tool for in vivo NIR imaging, as schematically shown in Scheme 1.
To synthesize the perylene monoimide-based NIR probe, we
started with commercially-available perylene-3,4,9,10-tetracarboxylic
dianhydride, which was converted to N-(2,6-di-isopropylphenyl)-9-
bromoperylene-3,4-dicarboximide (1) following a literature
Hyderabad 500007, India
b Academy of Scientific and Innovative Research (AcSIR), Ghaziabad 201002, India
c Department of Organic Synthesis and Process Chemistry, CSIR-Indian Institute of
Chemical Technology, Hyderabad 500007, India. E-mail: janaavijit2@gmail.com,
Avijit@iict.res.in
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9cc07322a
14182 | Chem. Commun., 2019, 55, 14182--14185
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
origin or genetic background.10 The tumor microenvironment is
mildly acidic due to the distorted metabolism, which is also known
as the ‘‘Warburg effect’’ of tumors and relative hypoxia.25 The
strong emission of the probe 2 in the pH range of 3.0–6.0, which
actually covers both normal and abnormal lysosomal pH, indi-
cated that the probe 2 can be an ideal candidate in the systematic
mapping of the cellular pH and thereby can be able to distinguish
cancerous cells over noncancerous cells.
Scheme 1 Schematic of the perylene monoimide (PeIm)-based NIR
probe for in vitro and in vivo bio-imaging.
Since the photophysical properties of the probe 2 revealed its
strong absorption and emission in the NIR region and high
sensitivity towards the solvent pH, it was employed for in vitro
imaging studies on the cancerous MCF7 cell line to understand its
potential in cell imaging (Fig. S2–S4, ESI†). In order to understand
the cytoplasmic distribution of 2 in detail, cells were counter-
stained separately with commercially available lysotracker green
(LTG) and mitotracker green (MTG), followed by recording images
using a confocal laser scanning microscope (CLSM). The counter-
staining experiment clearly revealed the co-localization of 2
(yellow fluorescence in Fig. 2(i)d) with LTG (Fig. 2(i)a–d and Fig.
S5, S6, ESI†), but not with MTG (Fig. 2(ii)a–d and Fig. S7, S8, ESI†).
These observations clearly signify that the probe 2 is exclusively
localized in the lysosomes of the MCF7 cell line. Furthermore, to
clarify whether the lysosomal localization of the probe 2 was its
exclusive property, similar subcellular localization experiments
with 2 were performed on a noncancerous CHO cell line following
a similar protocol to that used for the MCF7 cell line. As expected,
the probe 2 was internalized by the CHO cells (Fig. S9 and S10,
ESI†) and profoundly distributed in the lysosome (Fig. 2(iii)a–d,
yellow fluorescence in Fig. 2(iii)d and Fig. S11, S12, ESI†), but not
in the mitochondria (Fig. 2(iv)a–d and Fig. S13, S14, ESI†).
Furthermore, the quantitative colocalization studies conducted
by means of scatter plot and colocalization pixel map analysis
clearly demonstrate a profound localization of probe 2 in the
lysosome (Fig. 2(i)e and (iii)e) over mitochondria (Fig. 2(ii)e and
(iv)e). Although, the subcellular distribution of the probe 2 in both
procedure.24 Finally, 1 was coupled with 2-morpholinoethan-1-
amine to obtain the final NIR probe, N-(2,6-di-isopropylphenyl)-
9-(2-morpholinoethan-1-amine)perylene-3,4-dicarboximide (2) as
shown in Scheme S1 (ESI†).
After successful synthesis and characterization, the photophysi-
cal properties of probe 2 were investigated in detail in different
solvent systems. The absorption spectra of probe 2 in an acetoni-
trile solution (1 Â 10À5 M) revealed an intense absorption band
centered at 610 nm, while the emission maxima was red-shifted at
about 700 nm (Fig. 1a). When the emission spectra of 2 were
recorded in different solvents, a gradual bathochromic shift with
increased emission intensity on increasing the solvent polarity was
observed (Fig. 1b). This can be attributed to the suppression of the
photoinduced electron transfer (PET) from the morpholine N-atom
to the PeIm chromophore. In order to understand the effect of
H-bonding on the emission properties of the probe 2, emission
spectra were recorded in ACN/water binary mixtures. It was noted
that an increase in the percentage of water resulted in a further
bathochromic shift in the emission maxima to around 725 nm.
However, a gradual decrease in the emission intensity of the probe
2 (Fig. 1c) was observed because of the formation of aggregates. On
the other hand, when the emission spectra of 2 were recorded at a
different pH, it was observed that the probe 2 was less emissive
in neutral or in basic pH. However, on lowering the solution pH
(6.0–3.0), the emission intensity was greatly enhanced (Fig. 1d),
which suggested the protonation of the morpholine N-atom and
thereby the dissolution of the aggregates and also the complete
suppression of the PET process as mentioned earlier.
It has been well documented that a dysregulated pH is an
adaptive feature for most of the cancers irrespective of their tissue
Fig. 2 CLSM images of the MCF7 cell lines incubated with 2, DAPI, LTG
and MTG: (a) NIR channel, (b) blue channel, (c) green channel, (d) merged
images and (e) a colocalization pixel map of the corresponding NIR and
green channel. Insets of (ie–ive) are the corresponding scatter plots. (ie
and iie) Concentrated pixels along the diagonals indicate a high degree of
colocalization of probe 2 and LTG; (iie and ive): scattered pixels along the
diagonals indicate a lower degree of colocalization of probe 2 and MTG.
Scale bar: 25 mm.
Fig. 1 (a) Normalized absorption and emission spectra of 2 in ACN.
Emission spectra of 2 in (b) different solvents, (c) a binary mixture of
ACN and water and (d) in a different pH.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 14182--14185 | 14183

View Article Online
ChemComm
Communication
cancerous MCF7 cells and noncancerous CHO cells revealed a
profound accumulation in the acidic lysosomal compartment, the
most interesting observation was the emission signal from the
MCF7 cells (Fig. 2(i)a–d), which was remarkably high with respect
to the signal from the CHO cells (Fig. 2(iii)a–d). For the better
visualization of the pH-dependent colorimetric differences
between the cancerous MCF7 cells and noncancerous CHO cells, Fig. 4 CLSM images of MCF7 cells incubated with 2 and DAPI (ia–e) at pH
7.4. (iia–e) The same cell imaged after the buffer was changed to pH 4.5
3D Z-stack images with higher magnification are displayed in
Fig. 3, which clearly show a higher fluorescence intensity in the
with all the same microscopic parameters. Scale bar: 25 mm.
MCF7 cells compared to the CHO cells. Furthermore, the degree
of cellular uptake of 2 and thereby colorimetric differences in vitro
were also assessed quantitatively by means of flow cytometry
using both MCF 7 (Fig. 3c) and CHO (Fig. 3d) cell lines. Signifi-
cantly higher degrees of cellular uptake were observed over time
only for the cancerous MCF7 cells (Fig. 3c). This remarkably high
emission intensity in the MCF7 cells (Fig. 3a and c) can be
correlated to the inherent acidic nature of the cancerous cells
over the normal CHO cells (Fig. 3b and d). Therefore, the probe 2
can be an ideal diagnostic tool for the discrimination of cancerous
and noncancerous cells in vitro.
Now, the lysosome exclusive localization of the probe 2
could be attributed to either the exclusive physical localization
of 2 in the lysosomes or its presence in other organelles in the
endocytic pathway, but it was only emissive in the lysosome due
Fig. 5 CLSM images of (i) RKO, (ii) U87, (iii) CT26, (iv) GL261 and (v) HEK
293 cell lines incubated with probe 2, DAPI and LTG: (a) NIR, (b) blue
to the protonation of the morpholine N-atom at lower pH,
whereas in other organelles, it remained fluorescence silent.
To understand the 2nd postulate, i.e. the probe 2 may be
located in other organelles but was in the fluorescence OFF state,
MCF7 cells were first incubated with 2 in media at pH 7.4 for 4 h;
followed this, the cells were fixed and imaged using CLMS
channel, (c) merged image of (a and b), (d) green channel, (e) merged
images (a, b and d, and f) colocalization pixel map of the corresponding
NIR and green channel. Inset in (if–vf) is the corresponding scatter plot:
concentrated pixels along the diagonals indicate high degree of coloca-
lization of probe 2 and LTG. Scale bar: 25 mm.
(Fig. 4(i)a–e). Then, the medium was discarded and fresh medium (Fig. 5(v)a–e) cell line. The controlled imaging experiment clearly
of pH 4.5 (pH was adjusted using HCl) was added, and the cells shows a profound emission intensity for all the four cancerous
were further imaged with the same microscopic parameters after cell lines (Fig. 5(i–iv)a–e) in comparison to the normal HEK 293
20 min. The results revealed a significantly increased fluorescence cell line (Fig. 5(v)a–e).
within the cells (Fig. 4(ii)a–e and Fig. S17, S18, ESI†). The
We also tested the possibility of live cell imaging using the
increased fluorescence intensity at lower pH was because it forced probe 2 to establish it as a readily available bioanalytical probe
the probe 2 to be protonated within the organelles of a naturally for both in vitro and in vivo applications. MCF7 cells were pretreated
higher pH environment. This clearly shows that probe 2, which with 2 and LTG for 4 h. Cells were incubated at 37 1C under 5% CO2
was located in the lysosomes, was only in the fluorescent ON state and imaged at time points from 0 to 30 min using confocal
under normal cellular conditions.
microscopy. In the CLSM images (Fig. S15(i)a–e, ESI†), lysosomes
To explore the potential of the probe 2 in cancer diagnosis can be clearly seen as point-like fluorescent vesicles in the periphery
in vitro, controlled imaging experiments were performed on a set of the cell nucleus.
of four different cancer cell lines, namely, RKO (Fig. 5(i)a–e), U87
Images were taken continuously at 15 min intervals to track the
(Fig. 5(ii)a–e), CT26 (Fig. 5(iii)a–e), and GL261 (Fig. 5(iv)a–e) as fluorescent lysosomes over time, which showed the gradual move-
well as the normal Human embryonic kidney 293 (HEK 293) ment of the lysosomes towards the edge of the cell (Fig. S15(i)b–(iii)b,
ESI†), marked in the circle). It was also observed that the number of
fluorescent vesicles decreased over time with the appearance of
bigger and brighter vesicles, suggesting the fusion of several
lysosomes26 over the course of cellular events. Therefore, the
probe 2 was readily internalized by the living MCF7 cells, which
allowed us to visualize its real-time dynamics, which further
confirmed the applicability of probe 2 as an efficient diagnostic
tool for the evaluation of lysosomal health.
Fig. 3 CLSM images and flow cytometry data of the MCF7 and CHO cell
lines incubated with probe 2 and DAPI: (a) MCF7 and (b) CHO (all 3D Z
The advantages of a NIR fluorophore for in vivo imaging due
to the transparency of the tissues in the NIR window were
previously described. To test the in vivo imaging performance of
stack images). (c and d) Corresponding flow cytometry data in different
time intervals for the MCF7 and CHO cells, respectively.
14184 | Chem. Commun., 2019, 55, 14182--14185
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
also employed to visualize the real-time dynamics of lysosomal
movement in live MCF7 cells. Furthermore, it was successfully
used for in vivo deep-tissue imaging of a C57BL/6J mice model. The
present system could be a promising candidate for further bioana-
lytical, diagnostic and image-guided therapeutic development.
AJ and RM thank DST for the INSPIRE Faculty Grant (IFA15,
CH171, GAP0546), CA thanks SERB for the NPDF grant and VV
thanks CSIR for the fellowship. The CSIR-IICT Communication
number for this manuscript is IICT/Pubs./2019/317.
Fig. 6 In vivo NIR imaging of 2 in female C57BL/6J mice: (a) after 0 h, (b)
8 h, (c) 16 h, and (d) 24 h of i.v. of 2. (e) Ex vivo images of different organs
after 24 h. (f) % distribution of probe 2 in per unit mass of the respective
organs.
Conflicts of interest
2, an intravenous (i.v.) tail vein injection of 2 (4 mg kgÀ1) was
administered to 6–8 weeks old female C57BL/6J mice. Follow-
ing this, images were acquired at regular intervals over the
course of 24 h (Fig. 6a–d).
There are no conflicts to declare.
Notes and references
To have a better understanding of the biodistribution profile
of 2, mice were sacrificed after 24 h and different organs were
isolated and ex vivo imaged. The representative ex vivo images
showed the highest emission in the liver, followed by the
kidney and spleen (Fig. 6e), suggesting the gradual clearance
of the probe 2 from the body through urine over time. The
probe 2 was extracted in acidic isopropanol from different
organs at 24 h post i.v., and the biodistribution of 2 per unit
mass of the corresponding organs was determined. The results
showed that a major distribution of the administered probe 2
was observed in the spleen (460%), followed by the kidney
(430%) and liver (o10%) (Fig. 6f).
1 D. Cheng, J. Peng, Y. Lv, D. Su, D. Liu, M. Chen, L. Yuan and
X. Zhang, J. Am. Chem. Soc., 2019, 141, 6352.
2 D. Cheng, Y. Pan, L. Wang, Z. Zeng, L. Yuan, X. Zhang and
Y. T. Chang, J. Am. Chem. Soc., 2017, 139, 285.
3 Q. Gong, W. Shi, L. Li and H. Ma, Chem. Sci., 2016, 7, 788.
4 S. Luo, E. Zhang, Y. Su, T. Cheng and C. Shi, Biomaterials, 2011,
32, 7127.
5 A. Jana, L. Bai, X. Li, H. Ågren and Y. Zhao, ACS Appl. Mater.
Interfaces, 2016, 8, 2336.
6 R. Wang and F. Zhang, J. Mater. Chem. B, 2014, 2, 2422.
7 J. Fan, H. Dong, M. Hu, J. Wang, H. Zhang, H. Zhu, W. Sun and
X. Peng, Chem. Commun., 2014, 50, 882.
8 P. Wang, Y. Fan, L. Lu, L. Liu, L. Fan, M. Zhao, Y. Xie, C. Xu and
F. Zhang, Nat. Commun., 2018, 9, 2898.
9 C.-H. Tung, J. Qi, L. Hu, M. S. Han and Y. Kim, Theranostics, 2015,
5, 1166.
10 F. Bray, J. Ferlay, I. Soerjomataram, R. L. Siegel, L. A. Torre and
A. Jemal, Ca-Cancer J. Clin., 2018, 68, 394.
In order to validate the probe 2 to be applicable for practical
imaging, its in vivo toxicity was investigated on the C57BL/6J
mouse model. C57BL/6J mice (4–6 weeks old) were treated with 11 R. T. Greenlee, T. Murray, S. Bolden and P. A. Wingo, Ca-Cancer
2 at two different concentrations (4 and 16 mg kgÀ1). In
J. Clin., 2000, 50, 7.
12 Y. Kitagawa, S. Nishizawa, K. Sano, T. Ogasawara, M. Nakamura,
addition, in a control experiment, the mice were treated with
N. Sadato, M. Yoshida and Y. Yonekura, J. Nucl. Med., 2003, 44, 198.
Hanks’ balanced salt solution (HBSS) by i.v. injection every 13 L. Fass, Mol. Oncol., 2008, 2, 115.
14 A. Yuan, J. Wu, X. Tang, L. Zhao, F. Xu and Y. Hu, J. Pharm. Sci.,
alternative day for a period of 4 weeks. The weight and move-
ment of all the mice were checked daily. During the entire
2013, 102, 6.
¨
15 A. Nel, T. Xia, L. Madler and N. Li, Science, 2006, 311, 622.
treatment (4 weeks) with the probe 2, the body weight and 16 M. N. Moore, Environ. Int., 2006, 32, 967.
17 M. C. Mancini, B. A. Kairdolf, A. M. Smith and S. Nie, J. Am. Chem.
Soc., 2008, 130, 10836.
18 A. Hoshino, S. Hanada and K. Yamamoto, Arch. Toxicol., 2011, 85, 707.
movement of the mice were found to be similar for all of the
treated mice (data not shown) as well as the mice used for
the control experiment, which clearly shows that the corres- 19 A. M. Alkilany, P. K. Nagaria, C. R. Hexel, T. J. Shaw, C. J. Murphy
and M. D. Wyatt, Small, 2009, 5(6), 701.
20 A. Jana, K. S. P. Devi, T. K. Maiti and N. D. P. Singh, J. Am. Chem.
ponding doses of 2 were in a tolerable limit for all of the mice.
After the completion of the treatment, mice were sacrificed and
Soc., 2012, 134, 7656.
the tissues of five major organs, i.e., heart, kidney, liver, lung, 21 A. Jana, K. T. Nguyen, X. Li, P. Zhu, N. S. Tan, H. Ågren and Y. Zhao,
ACS Nano, 2014, 8, 5939.
22 H. Iwashita, S. Torii, N. Nagahora, M. Ishiyama, K. Shioji, K. Sasamoto,
and spleen, were collected and subjected to hematoxylin and
eosin (HE) staining, which showed no visible tissue necrosis for
S. Shimizu and K. Okuma, ACS Chem. Biol., 2017, 12, 2546.
all five major organs (Fig. S16, ESI†). Thus, the in vivo toxicology 23 C.-K. Tsung, X. Kou, Q. Shi, J. Zhang, M. H. Yeung, J. Wang and
G. D. Stucky, J. Am. Chem. Soc., 2006, 128, 5352.
24 K. M. Lefler, D. T. Co and M. R. Wasielewski, J. Phys. Chem. Lett.,
studies clearly demonstrate that the probe 2 could be considered
safe for future applications.
2012, 3, 3798.
In conclusion, we developed a perylene-derived highly 25 V. Estrella, T. Chen, M. Lloyd, J. Wojtkowiak, H. H. Cornnell, A. I.
Hashim, K. Bailey, Y. Balagurunathan, J. M. Rothberg, B. F. Sloane,
J. Johnson, R. A. Gatenby and R. Gillies, J. Cancer Res., 2013,
73, 1524.
photostable ultra-bright NIR probe, which is highly specific
for lysosome imaging in both fixed and live cells, and exploited
to discriminate cancer cells over the normal cells in vitro. It was 26 Q. Ba, G. Raghavan, K. Kiselyov and G. Yang, Cell Rep., 2018, 23, 3591.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 14182--14185 | 14185