Syntheses and photodynamic activity of some glucose-conjugated BODIPY dyes

Article abstract of DOI:10.1016/j.ejmech.2016.06.050

The syntheses of three water-soluble glucose-conjugated BODIPY dyes with different wavelength emissions and studies of their photodynamic therapeutic (PDT) action on human lung cancer A549?cell line are disclosed. Amongst the chosen compounds, the BODIPY

Full text of DOI:10.1016/j.ejmech.2016.06.050

Accepted Manuscript
Syntheses and photodynamic activity of some glucose-conjugated BODIPY dyes
Neelam Shivran, Mrityunjay Tyagi, Soumyaditya Mula, Pooja Gupta, Bhaskar Saha,
Birija S. Patro, Subrata Chattopadhyay
PII:
S0223-5234(16)30540-2
10.1016/j.ejmech.2016.06.050
EJMECH 8711
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 16 May 2016
Revised Date: 13 June 2016
Accepted Date: 28 June 2016
Please cite this article as: N. Shivran, M. Tyagi, S. Mula, P. Gupta, B. Saha, B.S. Patro, S.
Chattopadhyay, Syntheses and photodynamic activity of some glucose-conjugated BODIPY dyes,
European Journal of Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.06.050.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
+
hn
+
Apoptotic cells
A549 cells
ER-
Tracker
Overlay
BODIPY
The BODIPY-glucose conjugate localizes in endoplasmic
reticulum to exert photodynamic effect on A549 cells.

1
ACCEPTED MANUSCRIPT
Syntheses and photodynamic activity of some glucose-conjugated BODIPY dyes
#
#
#
Neelam Shivran, Mrityunjay Tyagi, Soumyaditya Mula, Pooja Gupta, Bhaskar Saha,
*
Birija S. Patro, and Subrata Chattopadhyay
Bio-Organic Division, Bhabha Atomic Resesarch Centre, Mumbai 400 085, India
E. Mail: schatt@barc.gov.in
#
Equal contributions
ABSTRACT- The syntheses of three water-soluble glucose-conjugated BODIPY dyes with
different wavelength emissions and studies of their photodynamic therapeutic (PDT) action
on human lung cancer A549 cell line are disclosed. Amongst the chosen compounds, the
BODIPY dye 4 possessing a glycosylated styryl moiety at the C-3 position showed best PDT
property against the A549 cell line. In particular, it induced reactive oxygen species-
mediated caspase-8/ caspase-3-dependent apoptosis as revealed from the increased sub G1
cell population and changes in cell morphology. These results along with its localization in
the endoplasmic reticulum, as revealed by confocal microscopy suggested that mitochondria
may not be directly involved in the photo-cytotoxicity of 4. Compound 4 did not induce any
dark toxicity to the A549 cells, and was non-toxic to normal lung cells.
Keywords: apoptosis; BODIPY dyes; lung cancer; photodynamic property.

2
ACCEPTED MANUSCRIPT
Highlights
•
•
•
Formulation of a BODIPY dye, possessing a glycosylated styryl moiety at the C-3
position as a potent photodynamic agent.
The BODIPY was phototoxic to human lung cancer A549 cell line, inducing caspase-
8
/ caspase-3-mediated apoptosis.
The BODIPY was non-toxic to normal human lung cell lines, and was devoid of any
dark cytoxicity.

3
ACCEPTED MANUSCRIPT
1
. Introduction
Photodynamic therapy (PDT) is a noninvasive treatment modality for a number of
diseases including certain types of cancers and pre-cancerous inductions [1-3], age-related
macular degeneration and actinic keratosis, localized infections, dermatological and
cardiovascular illnesses and wound healing [4-7]. It relies on using a photosensitizer (PS),
which on photo-excitation generates reactive oxygen species (ROS) in presence of oxygen to
achieve selective irreversible cell killing. It is particularly promising in the treatment of
multidrug-resistant (MDR) tumors, as both PS and light can be effectively localized in the
tumor [8]. Evidently the chosen PS plays a key role in the outcome of the PDT. Some of the
desirable attributes of the PS are its (i) ability to generate ROS, (ii) cellular internalization
and biodistribution, and (iii) long wavelength (preferably in the red and/ or near IR regions)
emission because of its better tissue permeability and cellular non-toxicity [9]. Long
wavelength light is preferred for living subjects because it causes less photo damage to cells,
and penetrates tissues better, while the higher wavelength light is absorbed by the tissues and
is converted to heat energy.
Selective destruction of malignant cells sparing the normal tissues is another
important issue in PDT. To a certain extent this is achieved by targeted light delivery to the
tumor site, and can be improved further by conjugating the PSs with low molecular weight
molecules that are better recognized by some of the overexpressed specific receptors in
tumor cells [10,11]. The biodistribution and cellular uptake of the PS depends on the balance
between its hydrophilicity and lipophilicity, as too high lipophilicity would hamper their
transport through blood vessel, while a high hydrophilicity would impede its cell membrane
penetration. Despite offering an alternative or complement to conventional therapies [12-14],
the clinical use of PDT is restricted due to non-availability of an ideal PS. A few
functionalized porphyrins, currently approved for clinical applications are not ideal PDT

4
ACCEPTED MANUSCRIPT
drugs due to their low extinction coefficients in the therapeutic window (650–800 nm), and
low absorptivity in mammalian tissues [15]. Therefore, there is a significant impetus to
develop novel and more efficient sensitizers for use in PDT.
The dipyrromethene-BF2 (4,4-difluoro-4-bora-3a,4a-diaza-s-indacene, BODIPY)
fluorophores have gained importance for diverse applications due to their advantageous
photo-physical properties including sharp absorption/ emission bands, high absorption
coefficients and excellent fluorescence quantum yields that are relatively insensitive to
changes in the local environment [16]. Most of these dyes have several characteristics of
ideal PDT agents such as low dark cellular toxicity, good cellular uptake, high extinction
coefficients, and low quantum yields for photobleaching. The BODIPYs are easy to
synthesize, and most importantly, are amenable to extensive synthetic modifications that can
help in tuning their florescence maxima and efficiency as well as ability to generate ROS. In
view of these, development of BODIPY-based PDT agents continues to be an area of intense
research, which has been elegantly reviewed [17,18]. Given that the BODIPYs are
lipophilic, attachment of hydrophilic entities such as polyethylene glycol chains and
carbohydrates to the BODIPY core might be useful in designing targeted probes for
diagnosis as well as PDT agents. The PSs with sugar moieties are of great importance
because sugar increases water solubility, membrane interaction and specific affinity for
malignant tumors [19]. Carbohydrate-functionalized macromolecules have shown promise in
photodynamic therapy and in vivo optical imaging due to their enhanced interactions with
tumor cell receptors via carbohydrate-mediated cell recognition processes, and increased
cellular uptake and solubility, without significant alterations of the physicochemical and
photophysical properties [20,21]. Earlier, BODIPYs containing various hydrophilic groups
have been used as imaging probes [22-24], and PDT agents [25-27]. Recently, cell
internalization of some BODIPY-based fluorescent dyes bearing carbohydrate residues have

5
ACCEPTED MANUSCRIPT
been studied by fluorescence microscopy [28,29], while synthesis and photophysical
properties of some BODIPY-conjugated iminosugar clusters [30] as well as of
lactosylceramide [31] have also been reported. In another study, some BODIPYs were
grafted on a carbohydrate platform to synthesize artificial light-harvesting antenna [32]. In
all these cases, the sugar moieties were attached to the dipyrrin carbon framework of the
BODIPY fluorophore. In a significant departure from these methods, an interesting synthetic
route for the direct attachment of the sugar moiety at the boron atom of the BODIPY
fluorophore has been described very recently [33].
For the past few years, we are actively pursuing research in chemistry and
applications of the BODIPYs [34-38]. The aim of the present study was to formulate some
new glycosylated BODIPYs that can be easily synthesized, are bioavailable, and show good
photosensitization to the target cells. Spin-coupling to heavy atoms (the ‘‘heavy atom
effect’’) that ensures good intersystem crossing (ISC) rates is often used to augment
generation of the cytotoxic ROS. However, the BODIPY derivatives are well-known to
1
generate copious amount of ROS, especially O that also leads to their oxidative
2
degradation [34,37]. Hence, we did not address this issue. Moreover, the BODIPY PSs with
heavy atoms, such as I, Br, Se, S etc. may offer increased dark toxicity, and altered bio-
distribution as well as pharmacokinetics [39,40].
Recently, a novel synthesis of several iodinated BODIPY-compounds, containing a
glucose moiety at the meso-appendage and oligo-ethylene glycol ethers at the 3,5-
disubstituted 4-hydroxystyryl groups has been patented. Based on the MTT-based
photocytoxicity data on MDA-MB-231 human breast cancer cell line, their potential
application in targeted cancer therapy has been proposed [41]. In another study, some
elegantly designed BODIPY molecules showed target-selective PDT property [42].
However, this warranted adoption of a complex synthetic strategy. Instead, we synthesized

6
ACCEPTED MANUSCRIPT
three glycoconjugated BODIPY dyes (2, 4 and 6) and compared their photosensitization
property with their non-glycoconjugated precursors, 1, 3 and 5 on the human lung carcinoma
A549 cell line. The commercially available BODIPY dye (PM567) was used as a reference
dye. The chemical structures of test samples are shown in Figure 1. In this panel of
molecules, we have incorporated a phenol moiety at the meso- or the pyrrole 3/5 positions of
the BODIPY core for the attachment of the glucose moiety. Introduction of the styryl phenol
moiety at the 3/5-position of the pyrrole ring also helped in synthesizing the dyes with red-
shifted emissions. Relevant physico-chemical parameters, such as absorbance/ emission
profiles, fluorescent quantum yields, and DNA binding characteristics of the dyes were also
determined. Next, the PDT property of the test compounds was examined and the results
mechanistically rationalized by assessing some biochemical parameters. We chose the
human lung cancer A549 cell line for the present studies as lung cancer is one of the most
common malignancy in humans and has become the leading cause of death. Despite the
rapid progress of surgery, radiotherapy, chemotherapy, and biotherapy, the long-term
survival rate of patients with lung cancer remains poor, and new therapeutic strategies are
urgently needed [43,44]. Since early lung cancer detection is now becoming feasible, PDT
may be applied to treat early stage lung carcinoma. Several clinical trials have established its
efficacy with superficial small tumors, while its use as a preoperative measure may reduce
tumor burden and the degree of surgery for larger tumors [45].
2
. Results and discussion
2
.1 Synthesis
The syntheses of the parent BODIPYs were straightforward. Thus, kryptopyrrole 7
was condensed with benzaldehyde or 4-hydroxybenzaldehyde in presence of para-toluene
sulphonic acid (PTS) to obtain compounds 8 [34] and 1 respectively. The reaction is usually
carried out in CH Cl when the BODIPYs are obtained in moderate yields with most of the
2
2

7
ACCEPTED MANUSCRIPT
aromatic aldehydes [34]. During the synthesis of 1, it was found that the reaction was very
slow and produced a dark brown mixture resulting in poor yield of the isolated product.
However, minor modifications of the reaction conditions such as use of THF as the solvent
and addition of DDQ at 0 °C improved the yield of 1, as the reaction was clean and the
product isolation easy. The BODIPY dyes containing C-3/5-CH substituted pyrrole rings
3
undergo the base-catalyzed Knoevenagel-type condensations with various aryl aldehydes.
This strategy has been extensively used to synthesize BODIPY derivatives with red-shifted
fluorescence [46,47]. Hence, for the present synthesis, the dyes, PM567 and 8 were
individually subjected to condensation with 1.0 equivalent of 4-hydroxybenzaldehyde in the
presence of piperidine and acetic acid to obtain the monostyryl derivatives 3 and 5
respectively (Scheme 1.) along with trace (5-7%) amounts of the corresponding distyryl
products, which were discarded by column chromatography.

8
ACCEPTED MANUSCRIPT
The required glycosylation of the phenolic functionality in 1 and 3 was achieved
using glucose pentaacetate as the glycosyl donor in presence of BF .Et O catalyst in CH Cl
2
3
2
2
to furnish 9 and 10 respectively (Scheme 2.). However, the same strategy with 5 led to its
considerable degradation, and furnished the required glycoside 12 in very low yield.
Glycosylation of 5 was subsequently achieved by the improved classical Konigs-Knorr
method, involving a base (K CO )-catalyzed reaction of 5 with 2,3,4,6-tera-O-acetyl-alpha-
2
3
D-glucopyranosyl bromide (11) in 1:1 H O/CHCl to obtain 12 [48]. The glucosyl donor 11
2
3
was synthesized from glucose pentaacetate by a tandem chemo-selective deacetylation of its
anomeric acetyl group and bromination with HBr-HOAc. It is well established that the
configuration of the anomeric sugar carbons remains unaffected during glycosylation [49].
Finally compounds 9, 10 and 12 were deacetylated with NaOMe in MeOH to obtain the
target BODIPYs 2, 4 and 6 respectively.

9
ACCEPTED MANUSCRIPT
2
.2 Photophysical characteristics

1
0
ACCEPTED MANUSCRIPT
The photophysical (absorption and fluorescence) data of the BODIPY-O-glycosides 2,
, and 6 in ethanol are summarized in Table 1. All the compounds exhibited typical spectral
4
characteristics of the BODIPY core with a narrow S →S absorption band with high molar
0
1
extinction coefficients, intense fluorescence emissions and small Stokes’ shifts (Figure SL1).
The absorption and emission spectra were almost mirror images of each other, indicating that
the emitting and absorbing species are similar. Compared to the non-styrylated BODIPY 2,
introduction of a styryl group as in 4 and 6 induced significant bathochromic shifts in the
absorption (51-63 nm) and emission (50-55 nm) maxima. The lower fluorescence quantum
yields of 4 and 6 indicated significant nonradiative energy loss due to rotation around the C-
Ar bonds [50]. Because the BODIPYs do not show solvatochromism, the photophysical
measurements were made in ethanol only.
Table 1. Photophysical parameters of the dyes 2, 4 and 6 in ethanol.
[a]
4
-1
-1 [b]
[c]
[d]
Dye
λ (nm)
abs
εmax (10 M cm )
λem (nm)
Φfl
[
e]
2
4
6
522.6
573.8
8.3
8.2
8.0
540
590
595
0.84
0.60
0.52
586.0
[
a]
[b]
[c]
[d]
Error: ± 0.2 nm. Extinction coefficients at the λ . Error: ± 1.0 nm. The fluorescence
max
quantum yields of the dyes 4 and 6 are relative to that of the dye rhodamine 101 (Φ = 1.0 in
fl
[
e]
EtOH). The fluorescence quantum yields of 2 is relative to that of PM567 (Φ = 0.84 in
fl
EtOH).
2
.3 DNA binding characteristics
By and large, the PSs exert their photo-dynamic action by generating O , which has
1
2
6
2
a short lifetime (ca. 2.0 µs) with a low apparent diffusion coefficient (4 × 10 cm s), and
thus, a very limited sphere of activity (about 150 nm in radius) in biological systems. This

1
1
ACCEPTED MANUSCRIPT
implies that binding to DNA is a desirable characteristic of the PDT agents, because the
1
generated O can induce DNA damage. Hence, the strong absorption bands of 2 and 4 at
2
5
22.6 nm and 573.8 nm respectively were conveniently used to determine their DNA
binding characteristics by spectrophotometric titrations. Incremental addition of double
stranded calf thymus (CT)-DNA (0–200 µM DNA base pair) to a fixed concentration of 2
(20 µM) and 4 (50 µM) led to gradual reductions in the intensities of their respective
absorption maxima (Figures SL2A and SL2B), confirming binding with DNA. Based on the
site-exclusion model [51], the equilibrium binding constants (K) in respective cases were
derived by quantitative analysis of the UV–visible data (Figures SL3A and SL3B). The
4
-1
4
-1
moderate K-values of 3.4 × 10 M and 1.8 × 10 M respectively for 2 and 4 suggested an
5
ionic binding, because the DNA-intercalators generally show higher binding constants ~10 -
7
-1
1
0 M . Moreover, well-known DNA intercalators such as coralyne produce new species on
binding with DNA [52]. Presently no additional absorption peak was observed on titration of
DNA with the dyes 2 and 4. The linear fits in regression analyses also indicated a single
mode binding of the dyes with DNA. We did not determine the DNA binding constants of
the other test samples based on the biological results (vide infra).
2
.4 Biology
For the biological studies, the in vitro photodynamic activities of 1-6 and the
commercial dye PM567 (as a reference molecule) were assessed by the MTT assay using the
highly invasive and metastatic human lung cancer A549 cell line. All the biological
experiments were carried out using the vehicle (≤ 0.1% DMSO in DMEM) that was non-
toxic to the cells. Throughout the manuscript, control refers to the photo-exposed cells in
2
vehicle. Initially the cells were photo-exposed (irradiance: 0.77 mW/cm ) for different
periods (0.5-4 h) in the presence of fixed concentrations (500, 1000 and 5000 nM) of 1-6 and
the growth inhibition was assessed by the MTT assay at 24 h following photo-exposure.

1
2
ACCEPTED MANUSCRIPT
At all the chosen concentrations of the compounds, photo-exposure time-dependently
reduced the viabilities of the A549 cells, with respect to the corresponding controls (Figure
SL4). However, the growth inhibitions at 1 and 2 h were not significantly different. Hence, a
8
0 min photo-exposure was chosen for the elaborate dose-dependent MTT assays with the
compounds 1-6 and PM567. Based on these results (Figure 2A), the growth inhibitory IC50
values, defined as the concentrations of the dyes required to kill 50% of the cells were
calculated and are shown in Table 2. In absence of light, none of the test compounds (up to
2
00 µM) showed any toxicity to the A549 cells, as revealed by the MTT assay results at 24 h
(data not shown). This excluded the possibility of any dark toxicity of the compounds.
Likewise, photo-exposure (up to 4 h) alone was non-toxic to the cells (Figure SL5),
establishing that light alone or temperature of the irradiating apparatus was not responsible
for the observed results. We used a long CFL tube (2 ft) along with a reflector for an uniform
photo-exposure. The irradiance, measured at 8-10 positions of the plates showed a marginal
(~2%) variation, confirming homogeneity of irradiation all over the plates.
The relative potency of the test compounds was 1>6~4>3~5~PM567>>2. The results
with the glycosylated dyes 4 and 6 vis-à-vis the corresponding parent dyes 3 and 5 clearly
revealed that glycosylation improved the potency of the styryl BODIPYs. But shifting the
emission maxima to longer wavelengths had insignificant effect on the PDT activity, as is
evident from the similar IC (~6.5 µM) values of the non-conjugated PM567 dye and the
5
0
conjugated styryl BODIPYs 3 and 5. Earlier we have shown that presence of a bulky meso-
1
substitution such as an aryl group in the BODIPYs reduce their O generating ability [34]. It
2
is possible that the large glycosylated phenol substitution at the meso-position of compound 2
1
does not allow significant O generation, explaining its poorer PDT activity compared to that
2
of its non-glycosylated precursor 1. This is consistent with a previous report, wherein the in
vitro and in vivo photosensitizing property of some lactose-conjugated porphyrinoids

1
3
ACCEPTED MANUSCRIPT
depended on the site of the sugar in the PSs [53]. It is worth noting that earlier, compound 1
showed impressive PDT property against the human colon carcinoma cell line HCT116 [54].
Table 2. Comparative cytotoxicities of the chosen BODIPYs against A549 human lung
a
cancer cells
a
Test compounds IC (µM)
5
0
*
*
PM567
6.8 ± 1.8
2.1 ± 0.6
> 10
1
2
3
4
5
6
*
*
*
*
6.5 ± 2.1
2.7 ± 0.8
6.4 ± 2.0
2.5 ± 1.1
*
*
a
The IC values were calculated using the MTT data shown in Figure 2A. The experiments
5
0
were repeated three times with similar results. All determinations were made in four
*
**
replicates, and the values are means ± S. E. M. P<0.001 compared to 2, P<0.001 compared
to the corresponding non-glycosylated BODIPYs.
Despite showing the best result amongst the test compounds, the photo-cytotoxicity of
1
was primarily due to necrosis as revealed by phase contrast microscopy of the trypan blue-
stained cells. Photo-exposure of the cells, treated with 1, but not with 4 and 6 showed
significant accumulation of the blue dye in the cells within 2 h (Figure 2B). This indicated
extensive membrane damage induced by 1, leading to necrosis. Hence we excluded
compound 1 for further studies. Although both 4 and 6 showed similar potency, compound 4
was easily accessible from commercially available materials. Hence, the subsequent

1
4
ACCEPTED MANUSCRIPT
mechanistic studies were carried out with the BODIPY dye 4. However, the cellular uptake
and distribution studies were carried out with all the chosen compounds 1-6 and PM567.
Apoptosis induction is a preferred mode of killing the cancer cells due to less side
effects and immune reactions [55]. In the present study, apoptosis induction by 4 was
confirmed by following several apoptosis-specific parameters: (i) sub-G1 cell population, (ii)
cell morphology, and (iii) cell survival analyses in the presence of a pan-caspase (Z-VAD-
FMK) as well as three specific caspases inhibitors. At a 30 min photo-exposure, compound 4
(5-50 µM) increased the sub-G1 cell population by ~2.8-3.3 folds, compared to control. The
effect was more severe with a 80 min photo-exposure, when the sub-G1 cell population
increased by 5-6 folds (Figures 3A and 3B). There was no alteration in the sub-G1 population
in the cells treated with 4 (up to 50 µM) in the absence of any photo-exposure. This clearly
established the lack of dark toxicity of 4 against the A549 cells.
The phase contrast microscopy, performed to examine the effect of compound 4 (up
to 2.5 µM) in conjunction with a 80 min photo-exposure showed obvious and significant
morphological changes in the A549 cells. At 24 h after the treatment, the cells showed
marked changes in the cell outline, with irregular disruptions in the optical diffraction halo,
and the effect was concentration dependent. The number of shrinking cells or cells with
blebbing membranes was notably increased in the 4-treated cells than in the control (Figure
3
C). There were number of floating cells without any adhering ability in the treated group
than in the controls.
Apoptosis can be induced by the ligation of plasma membrane death receptors, which
stimulate the ‘extrinsic’ pathway, or by perturbation of intracellular homeostasis, the
intrinsic’ pathway [56,57]. The extrinsic pathway is initiated by the binding of a member of
the tumor necrosis-factor (TNF)-family of death-receptor ligands to their cognate receptors
TNFR or Fas). The oligomerized receptors subsequently recruit adaptor protein (TRADD or
‘
(

1
5
ACCEPTED MANUSCRIPT
FADD), which in turn, recruits pro-caspase-8 and/or pro-caspase-10 to form the death
inducing signaling complex (DISC). As a part of the DISC, pro-caspase-8 becomes
autocatalytically activated to cleave the effector pro-caspase-3/7. The intrinsic pathway is
linked primarily to mitochondrial changes, directly inducing the release of cytochrome c into
the cytosol and apoptosome complex formation with activation of caspase-9. Both pathways
converge on caspase-3 activation, resulting in nuclear degradation and changes in cellular
morphology. In our results, treatment of the A549 cells with 4 (2.5 µM) followed by a 80 min
photo-exposure stimulated the activities of caspase-3 (~2.9 fold) and caspase-8 (~6.2 fold)
without any increase in the caspase-9 activity, compared to the untreated control cells (Figure
4
A). The specific inhibitors (each 20 µM) for caspase-3 (Z-DEVD-FMK) and caspase-8 (Z-
IETD-FMK) abrogated such activation of the respective caspases by 46.7% and 48.6%.
Expectedly, the caspase-9 specific inhibitor (Z-LEHD-FMK, 20 µM) was ineffective (data
not shown). To further confirm the involvement of caspase-8/caspase-3 pathway in the photo-
toxicity of 4 in the A549 cells, we assessed the cell survival (at 24 h) in the absence and
presence of the above caspase-specific inhibitors as well as the pan-caspase inhibitor (each,
2
0 µM). Pre-incubation of the cells with the caspase-8, caspase-3 as well as the pan-caspase
inhibitors increased their viability by 50.6%, 36.5% and 77.2% respectively, compared to the
only compound 4-treated and 80-min photo-exposed cells. However, the caspase-9 inhibitor
did not show significant effect on cell survival (Figure 4B). None of the above inhibitors, on
their own altered the activities of the caspases and the cell survival in absence of 4. Our
immunoblots (Figure 4C) also showed a time-dependent reduction in the pro-caspase-3 levels
in the compound 4 plus 80-min photo-exposed cells compared to the control cells. Taken
together, these results suggested involvement of the extrinsic apoptotic pathway in the photo-
toxicity of 4 to the A549 cells, where caspase-8/caspase-3 activation plays a major role. No
caspase-3 activation was noticed by treating the cells with 4 alone (data not shown). Current

1
6
ACCEPTED MANUSCRIPT
evidence suggests that the mitochondrial pathway is most common in PDT-induced apoptosis
in cells. However, the caspase-8-dependent pathways may also be important, when the
dominant pathway is suppressed [58].
Due to their higher metabolic rates, cancer cells are generally under more oxidative
stress than the normal cells [59]. PDT causes photooxidative damage to proteins and lipids
that reside within a few nanometers of the PSs binding sites [58]. There is strong evidence
that ROS can induce the apoptotic process [60]. Hence, we examined if ROS generation is
responsible for the photo-toxicity of 4, by carrying out the sub-G cell population assay in the
1
1
absence and presence of β-carotene (a specific O -inhibitor) [61] and N-acetyl cysteine
2
(NAC, a cell-permeable antioxidant). In addition, the cell survival assay was also carried out
under similar conditions in the presence of increasing concentrations of β-carotene. As
revealed from the results (Figures 5A and 5B), incubation of the cells with β-carotene (50 and
1
00 µM) or NAC (5 mM) along with 4 (2.5 µm) followed by a 80 min photo-exposure
reduced the sub-G cell population by 24.2%, 36.7% and 32.4% respectively, compared to the
1
only 4-treated and 80-min photo-exposed cells. In the MTT assay, β-carotene dose-
dependently reduced the photo-toxicity of 4, offering 46.5% protection at a concentration of
1
5
0 µm. The above results confirmed that the intracellular ROS, especially O was
2
responsible for the PDT property of 4 in the A549 cells.
Photodynamic efficacy is principally determined by the subcellular localization of a
PS that governs the primary site of photodamage [62]. Some PSs show a broad distribution,
while some may localize more specifically. Hence, fluorescence microscopic studies were
carried out to investigate the cellular uptake and localization of the test compounds. All the
BODIPY dyes exhibited cytoplasmic localization in the A549 cells as revealed from the red
fluorescence (Figures 6A and 6B, middle panel). The images of the cells, stained with the
nucleus specific dye, Hoechst-33342 are shown in the left panel of the same figures. The

1
7
ACCEPTED MANUSCRIPT
merged images (right panel) clearly revealed that the intra-cellular accumulation of all the
dyes was outside the nucleus. Compared to the other dyes, PM567 showed lower
fluorescence, indicating its less cellular uptake. Also, the patchy fluorescence in the cells,
stained with 1 indicated its non-uniform presence in the cytoplasm with higher accumulation
near the membrane. This may also account for its observed necrotic property. The other dyes
had uniform florescence all over the cytoplasm of the cells.
The subcellular localization of 4 was ascertained by confocal microscopy using dual
staining with 4 (100 nm) followed by the organelle-specific fluorescence probes viz. Lyso-
Tracker Green, Mito-Tracker Green or ER-Tracker Green. The clinically used PS, photofrin
has been shown to concentrate into plasma membranes or cytoplasm upon brief incubation,
and in the Golgi complex or ER upon prolonged incubation [63,64]. In our experiments, fast
cellular internalization of 4 was noticed within a few minutes. Although we have incubated
the cells with 4 up to 4 h, no significant change in its distribution in the subcellular
compartments was noticed by varying the incubation time points (0.5, 1, 2 and 4 h).
Presently, the confocal images for 1 h-incubation with 4 is shown in Figure 7, as the same
incubation time was used for the experiments, prior to photo-exposure. The left and middle
panels show the images of the cells, loaded with the respective fluorescence probes and
compound 4, while their superimpositions are shown in the right panel. There was an almost
identical overlap of the images of the cells loaded with compound 4 and the ER-Tracker,
indicating that ER is the preferential accumulation site of compound 4. In comparison,
colocalization of compound 4 was only partial with Mito-Tracker and very less with Lyso-
Tracker. This result is consistent with the fact that compound 4 did not activate caspases-9 to
induce mitochondrial apoptosis in the A549 cells. Several observations have indicated that
the PSs that localize within mitochondria or ER promote apoptosis. Since compound 4 did
not accumulate in the plasma membrane or lysosomes, it did not promote cell necrosis

1
8
ACCEPTED MANUSCRIPT
[
58,65]. It is also very unlikely to directly damage DNA, as the nucleus remained free of
compound 4. No significant alteration in the morphology was observed on incubating the
cells with 4 up to 4 h in dark (data not shown).
Activation of the caspases is perhaps the most well characterized apoptotic cascade,
and caspase-3 activation is a landmark event in apoptosis [66]. Our results showed activation
of caspase-8 and caspase-3 in the A549 cells by 4 under photo-exposure leading to increased
sub-G1 cell population that was abrogated by the specific as well as pan-caspase inhibitors. It
has been reported that procaspase-3 localizes in the cytoplasm and that caspase-3 activation is
initiated in the cytosol, followed by its redistribution to the nuclear compartment [67,68].
Because compound 4 also accumulates in the cytoplasm of the cells, it is expected to induce
the caspase-3-mediated apoptosis. Structurally, the BODIPYs constitute half of the
porphyrins that are known to be specifically accumulated in tumor cells. Hence, we
anticipated good tumor-selectivity of the BODIPYs. This was confirmed by the MTT assay
using normal human lung cell lines, WI-38 and L132. Compound 4 (50 µM) induced only 5-
7
% cytotoxicity in these cells under photo-exposure.
The main objective for adding a sugar moiety to the BODIPY scaffold was to ensure
preferential accumulation and/or retention in cancer vis-à-vis normal cells. To probe these,
we studied the uptake and retention of the dyes 3 and 4 (2.5 µM each) in A549 lung cancer
cell line from their fluorescence, using flow cytometry. The dye 4 is the glycosylated
derivative of dye 3. The relative amounts of the dyes inside the cells were quantified in terms
of the mean fluorescence intensity (MFI) in the red (FL3) channel [69]. A higher uptake and
longer retention of the dye 4 in the cells was observed, compared to 3 (Table 3). This may be
one of the reasons of its enhanced capacity in inducing death in the cancer cells vis-à-vis 3.
To rationalize the selective photocytotoxicity of 4 to cancer cells over normal cells, we also
carried out similar experiments with L132 and A549 cell lines. The initial uptake (at 2 h) of 4

1
9
ACCEPTED MANUSCRIPT
by A549 cells was significantly higher than by L132 cells. Since the internalization of 4 was
very fast and the experiments were carried out within 80 min, its selective photocytotoxicity
to the cancer cells compared to normal cells is expected. The higher uptake of 4 by the A549
cells compared to the L132 cells may be due the altered expression levels of many of the
plasma membrane proteins in the cancer cells vis-à-vis normal cells [70].
Table 3. Comparative uptake and retention of 3 and 4 in human lung cancer A549 cells and
a
of 4 in normal (L132) vs cancer (A549) human lung cells
Sample
MFI in A549 cell line
Comp. 3 Comp. 4
MFI of compound 4
L132
A549
Control
2.4±1.1
4.6±1.2
5.4±1.5
*
*,#
*
*,τ
2
h
88.0±6.8
98.5±7.2
63.5±4.5
95.3±6.3
66.2±4.6
*
*
*
*
#
#
#
#
*
*
*
*,$
4
8
h
h
53.7±6.2
43.9±5.5
40.9±5.9
38.4±4.8
68.9±6.3
58.4±6.1
52.5±5.8
46.7±5.9
56.8±5.4
49.5±5.8
41.5±4.2
*
49.4±6.0
41.6±3.7
*
1
2
6 h
4 h
a
The A549 cells were treated with 3 or 4 (2.5 µM) for different periods and the intracellular
concentrations of the compounds were analyzed by flow cytometry from the red florescence
FL3 channel). The data are presented in terms of MFI in arbitrary units. Similar experiments
(
were also carried out with 4 (2.5 µM) using A549 and L132 cells. The experiments were
repeated three times with similar results. All determinations were made in four replicates, and
*
#
the values are means ± S. E. M. p<0.001 compared to control cells, p<0.05 compared to the
$
τ
3
-treated cells; p<0.05, p<0.01 compared to the L132 cells.

2
0
ACCEPTED MANUSCRIPT
We adopted indirect methods to ascertain apoptosis induction and ROS production by
the dye 4 under photo-exposure. This is because of the overlapping fluorescence spectra of
the respective probes (FITC tagged annexin-V and DCFDA respectively) with that of 4 at its
1
IC concentration. For the same reason, we did not measure the amount of O produced by
5
0
2
it, by the 1,3-diphenylisobenzofuran bleaching method [71]. Nevertheless, we have provided
multiple lines of evidence to ascertain that the photo-toxicity of 4 to the A549 cells follows
apoptosis by assessing the sub G1 cell population, cell morphology and identifying the
involvement of the caspase-mediated pathway. We have also used two intracellular ROS
1
inhibitors, including β-carotene, a well-known O -specific inhibitor (at different
2
1
concentrations) to provide ample evidence of the involvement of O in the process.
2
3
. Conclusions
Overall, we developed a new water-soluble BODIPY dye 4 possessing a glycosylated
styryl moiety at the C-3 position by a simple synthetic route. The dye 4 showed impressive
photo-sensitization property against the human lung carcinoma (A549) cell line, inducing
ROS-mediated apoptosis via the caspase-8/caspase-3-dependent extrinsic pathway. Its
accumulation in cell cytoplasm matched with the localization site of procaspase-3 to ensure
apoptosis without any necrosis. The dye 4 did not show any dark toxicity to human normal
lung (WI-38 and L132) and cancer (A549) cell lines, and was non-toxic to the normal lung
cells even on photo-exposure. Compared to dye 4, some halogenated BODIPYs earlier
showed lower IC -values in their photo-dynamic property [17,18]. However, their photo-
5
0
toxicity against the non-targeted normal cells has not been verified. Considering their
possible dark toxicity, increasing attention is being paid to develop halogen-free BODIPYs as
PDT agents, and BODIPYs with extended conjugation have emerged as promising candidates
[
17,39,40]. However, it is difficult to synthesize these BODIPYs in sufficient amounts for
their clinical applications. By contrast, dye 4 can be easily synthesized in an overall yield of

2
1
ACCEPTED MANUSCRIPT
1
1.6% in 3-steps from the commercially available PM567 dye using inexpensive
chemicals/reagents, is bioavailable, and shows selective photosensitization to the target cells.
4
. Experimental Section
4
.1 Synthesis
The syntheses of the target BODIPYs were achieved as shown in Scheme 1 and
Scheme 2. Detailed experimental procedures as well as the characterization data of the
BODIPYs and the intermediates involved in the syntheses are provided in the supplementary
material.
4
.2 General Biology
Penicillin,
streptomycin,
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT), β-carotene, N-acetylcysteine (NAC), Triton X-100 (0.1%), Tween 20,
propidium iodide (PI), phenylmethylsulfonyl fluoride (PMSF), aprotinin, leupeptin, sodium
orthovanadate (NaVO ), and rabbit polyclonal antibody to caspase-3, RNase were procured
4
from Sigma chemicals (St. Luois, MO). Other chemicals used were: Dulbecco’s modified
Eagle’s medium (DMEM), fetal bovine serum (FBS), trypan blue (Gibco Life Technologies,
Carlsbad, CA), Hoechst 33342 (Molecular Probes, Inc., Eugene, OR), pan caspase inhibitor
(Z-VAD-FMK), specific inhibitors of caspase-9 (Z-LEHD-FMK), caspase-3 (Z-DEVD-
FMK) and caspase-8 (Z-IETD-FMK) (R&D Systems, Minneapolis, MN), and rabbit
PLUS
polyclonal antibody to β-actin (Abcam, Danvers, MA). Lumi-Light
western blotting kit
PLUS
and cell death detection
kit were procured from Roche Applied Science (Baden-
Wurttemberg, Mannheim), while nitrocellulose membrane (BioTrace® NT) was from Pall
Life Sciences (Easthills, NY).
4
.2.1 Cell culture

2
2
ACCEPTED MANUSCRIPT
The A549 cell line, procured from National Centre for Cell Science, Pune, India were
cultured in DMEM medium, supplemented with 10% FBS, 100 U/mL penicillin and 100
o
µg/mL streptomycin. The cells were grown at 37 C under an atmosphere of 5% CO .
2
4
.2.2 Drugs preparation
Stock solutions (50 mM) of the test samples in DMSO were reconstituted in DMEM
medium without any nutrient to attain the required doses of them for the MTT assay. For the
other assays, the samples were prepared as above using 5 mM stock solutions of the
respective compounds. The final DMSO concentration in all the samples was <0.1%.
4
.2.3 Photo-exposure protocol
The cells were irradiated with white light from a CFL source (2 ft long, 20 W,
Phillips), mounted on polished reflector unit. The distance between the light source and the
surface of the irradiated solutions was ~5 cm. The irradiance on the plates was measured at 8-
1
0 locations of the plates with a model LX-101A research light meter (Eztech Instruments,
2
Nashua, NH, USA). The mean value was 0.77 mW/cm (variation ~2%). The photo-
irradiation unit was housed in the CO atmosphere.
2
4
.2.4 MTT assay
Viabilities of the control and treated cells were determined by the MTT reduction
4
assay [72]. The A549 cells (1 × 10 /well) were grown in 96-well plates in DMEM medium
supplemented with 10% FBS at 37 °C in a 5% CO atmosphere. Cells were plated overnight
2
prior to any treatment. The unsynchronized cells were treated with vehicle (0.1% DMSO) or
the compounds (each 500, 1000 and 5000 µM) for 1 h in serum-free medium, washed twice
with phosphate-buffered saline (PBS), followed by photo-irradiation for different periods
(
0.5, 1, 2, and 4 h) at 37 °C in a 5% CO atmosphere. The cells were subsequently incubated
2
in a regular medium for 24 h, washed once with PBS, MTT solution (0.5 mg/mL, 100 µL)
o
was added to each well and kept at 37 C for 6 h. The formazan crystals in the viable cells

2
3
ACCEPTED MANUSCRIPT
were solubilized with 0.01 N HCl (100 µL) containing 10% SDS and the absorbance at 550
nm read.
For determining the IC -values, the assay was carried out as above, using different
5
0
concentrations of the compounds and a 80-min photo-exposure. Similar assays were also
carried out by pre-incubating the cells with the specific and pan caspase inhibitors (each
2
0 µM) or increasing concentrations of β-carotene (0-100 µM) for 1 h prior to the addition of
the test compound 4 (2.5 µM). The dark toxicities of the compounds were assessed at 48 h as
above, omitting any photo-exposure.
4
.2.5 Phase contrast microscopy
5
The A549 cells (1 × 10 cells/well) were cultured overnight on cover slips in 6-well
plates. The cultures were incubated with vehicle (0.1% DMSO) or the dyes 1, 4 and 6 (each
.5 µM) for 1 h, followed by photo-irradiation for 80 min. After addition of serum, the cells
2
were incubated for 2 h, washed twice with PBS, trypan blue (0.1% in PBS) added and
incubated further for 5 min. The cells were washed once with PBS, mounted on glass slides,
visualized, and representative fields of the cells were photographed using an Axioskop II Mot
plus (Zeiss) microscope (40 × objective, 0.65 Ph2), fitted with a Axiocam MRc camera.
Similar experiments were also carried out after treating the cells with different concentrations
of 4 as above, and visualizing the cells after 24 h without trypan blue staining.
4
.2.6 Flow cytometry
The hypodiploid DNA content were analyzed as a marker for apoptosis by flow
cytometry, after staining with PI. The cells were incubated with 4 (0-50 µM) for 1 h, washed
two times with PBS, DMEM (200 µL) added, and subsequently exposed to light for different
time periods (30 and 80 min). After addition of serum, the cells were incubated for 24 h,
washed once with cold PBS, incubated with PI (400 µg/mL) and RNAse A (200 µg/mL) in 1
o
mL hypotonic buffer (0.1% sodium citrate plus 0.1% Triton X-100) for 30 min at 37 C, and

2
4
ACCEPTED MANUSCRIPT
analyzed with a Partec CyFlow® Space flow cytometer using the FlowJo program. Cellular
debris was excluded from the analyses by raising the forward scatter threshold, and the DNA
4
content of the nuclei was registered on a logarithmic scale. At least 2 × 10 cells of each
sample were analyzed. The cell number read by flow cytometry for every single sample was
same, although the total areas under some curves look different. This is because of the
normalization done by the software in the overlayed histograms. The normalization is
required to avoid disproportionate heights of the histograms of different fluorescence
intensities for the two overlaid samples, and is achieved by plotting "% of max" instead of
"total cells counted or events taken" in the Y-axis [73]. The apoptotic nuclei appeared as
broad hypodiploid DNA peaks. Similar experiments were also carried by treating the cells
with NAC (5 mM) and β-carotene (50 and 100 µM) along with compound 4 (2.5 µM) and
photo-exposure for 80 min.
5
For the dye retention studies, the A549 cells (1×10 per well), grown in 6 well plates
as above were washed twice with PBS and treated with vehicle (0.1% DMSO) or the dyes, 3
and 4 (2.5 µM each) for 2 h in a serum-free medium. After washing twice with PBS, the cells
were incubated in a regular medium at 37 °C in a 5% CO atmosphere for different periods as
2
described in the Table 3. The cells were trypsinized, washed with PBS and analyzed through
flow cytometry. Similar experiments were also carried out using dye 4 and L132 cells. The
data are presented in terms of MFI.
4
.2.7 Fluorescence microscopy
A549 cells seeded in 6-well plates on coverslips were loaded with the BODIPY dyes
(1 µM) for varying periods (0.25, 0.5, 1, 2 and 3 h) at 37 ºC, washed with PBS, subsequently
stained with Hoechst 33342 (10 µM), washed once again with PBS, mounted with 70%
glycerol, and analyzed under an Axioskop II Mot plus (Zeiss) microscope (40 × objective,
0
.65 Ph2). Some adjustments to the stretch and tone of both blue and red images were made

2
5
ACCEPTED MANUSCRIPT
while overlaying the images. This was done only for better contrast and to ensure that the
blue color of the Hoechst-stained nucleus and red color of the dye are both visible.
4
.2.8 Confocal microscopy
The overnight-seeded cells were washed twice with PBS, stained with dye 4 (100 nM
for 2 h), washed again with PBS, and then stained with LysoTracker Green (500 nM),
MitoTracker Green (100 nM), or ER-Tracker Green (500 nM). The cells were incubated for
3
0 min at 37 °C. In all cases, the cells were washed again with PBS before imaging through
confocal microscope (LSM 780, ZEISS). An excitation wavelength of 488 nm and emission
range 500-530 nm were used for visualization the Mito-, ER- and Lyso-Tracker spots. For
visualization of 4, excitation at 543 nm and emission at 560-640 nm were used. The
concentration of 4 was chosen to exclude any absorption of the dye in the green channel.
4
.2.9 Caspases activity assays
The assays were performed with a caspase-3 colorimetric kit or caspase-8 and
caspase-9 fluorimetric kits according to the manufacturer’s protocol. Briefly, cells (1 ×
6
1
0 /well), seeded in 90 mm plates were incubated with 4 (2.5 µM) for 1 h followed by
photo-illumination for 80 min, and the individual caspase activities were assayed at 16 h. The
untreated but photo-exposed cells served as the control. Similar experiments were also
carried out by pre-incubating the cells with the inhibitory peptides, Z-VAD-FMK, Z-IETD-
FMK, Z-LEHD-FMK (each 20 µM), prior to other treatments.
4
.2.10 Western blots
The whole cell extracts were prepared by lysing the 4 (0 or 2.5 µΜ)-treated and
6
photo-exposed (80 min) A549 cells (1 × 10 /well) in a lysis buffer (20 mM Tris (pH 7.4), 250
mM NaCl, 2 mM EDTA (pH 8.0), 0.1% Triton X-100, 0.01 µg/mL aprotinin, 0.01 µg/mL
leupeptin, 0.4 mmol/L PMSF, and 4 mmol/L NaVO ). The lysates were spun at 16500 × g for
4
o
1
0 min, the supernates collected and kept at -70 C. The whole cell lysates (30 µg) were

2
6
ACCEPTED MANUSCRIPT
separated by 12% SDS-polyacrylamide gel electrophoresis, and electrotransferred to
nitrocellulose membrane. The membranes were blocked for 1 h at room temperature in TBST
buffer (10 mM Tris–HCl, pH 8.0, 150 mM NaCl, and 0.1% Tween-20) containing 5% (w/v)
o
nonfat milk, and then incubated overnight at 4 C with the Caspase-3 specific primary
antibody. After several washes, HRP-conjugated secondary antibody was added, the
membranes were incubated further for 1 h, and the blots were developed using a Lumi-
PLUS
Light
western blotting kit. The bands were detected using a Kodak Gel-doc software and
the intensity ratios of immunoblots to that of normal control, taken as 1 (arbitrary unit) were
quantified after normalizing with respect to the loading controls.
4
.3 Statistical analyses
The data were analyzed using a paired t test for the paired data or one way analysis of
variance (ANOVA) followed by a Dunnet multiple comparisons post test. Values are
expressed as means ± S.E.M. A probability value of P<0.05 was considered significant.

2
7
ACCEPTED MANUSCRIPT
References
[
[
[
[
1]. R. Bonnet, Chemical Aspects of Photodynamic Therapy; Gordon and Breach Science:
Amsterdam, 2000.
2]. T.J. Dougherty, An update on photodynamic therapy applications. J. Clin. Laser Med.
Surg. 20 (2002) 3–7.
3]. D.E. Dolmans, D. Fukumura, R.K. Jain, Photodynamic therapy for cancer. Nat. Rev.
Cancer 3 (2003) 380–387.
4]. P. Babilas, S. Schreml, M. Landthaler, R.M. Szeimies, Photodynamic therapy in
dermatology: state-of-the-art. Photodermatol. Photoimmunol. Photomed. 26 (2010)
1
18–132.
[
[
[
5]. S. Kossodo, G.M. LaMuraglia, Clinical potential of photodynamic therapy in
cardiovascular disorders. Am. J. Cardiovasc. Drugs 1 (2001) 15–21.
6]. G. Jori, Photodynamic therapy of microbial infections: state of the art and
perspectives. J. Environ. Pathol. Toxicol. Oncol. 25 (2006) 505–519.
7]. A.F. Cruess, G. Zlateva, A.M. Pleil, B. Wirostko, Photodynamic therapy with
verteporfin in age-related macular degeneration: a systematic review of efficacy,
safety, treatment modifications and pharmacoeconomic properties. Acta Ophthalmol.
8
7 (2009) 118–132.
[
[
[
8]. M.A.M. Capella, L.S.J. Capella, A light in multidrug resistance: photodynamic
treatment of multidrug-resistant tumors. J. Biomed. Sci. 10 (2003) 361–366.
9]. B. Ballou, L.A. Ernst, A.S. Waggoner, Fluorescence imaging of tumors in vivo. Curr.
Med. Chem. 12 (2005) 795–805.
10]. W.M. Sharman, J.E. van Lier, C.M. Allen, Targeted photodynamic therapy via
receptor mediated delivery systems. Adv. Drug Delivery Rev. 56 (2004) 53−76.

2
8
ACCEPTED MANUSCRIPT
[
11]. G. Zheng, H. Li, M. Zhang, S. Lund-Katz, B. Chance, J.D. Glickson, Low-density
lipoprotein reconstituted by pyropheophorbide cholesteryl oleate as target-specific
photosensitizer. Bioconjugate Chem. 13 (2002) 392−396.
[
[
[
12]. S.B. Brown, E.A. Brown, I. Walker, The present and future role of photodynamic
therapy in cancer treatment. Lancet Oncol. 5 (2004) 497–508.
13]. R.R. Allison, G.H. Downie, R. Cuenca, X.H. Hu, C.J. Childs, C.H. Sibata,
Photosensitizers in clinical PDT. Photodiagnosis Photodyn. Ther. 1 (2004) 27–42.
14]. T. Hasan, B. Ortel, N. Solban, B.W. Pogue, Photodynamic therapy of cancer. In:
D.W. Kufe, R. Bast, W. Hait, W.K. Hong, R.E. Pollock, R.R. Weichselbaurm, R.C.
Bast, Jr., T.S. Gansler, J.F. Holland and E. Frei, III, editors. Cancer Medicine, 7th ed.
Hamilton, Ontario, Canada: BC Decker, Inc.; 2006. p. 537–48.
[
15]. A.E. O’Connor, W.M. Gallagher, A.T. Byrne, Porphyrin and nonporphyrin
photosensitizers in oncology: preclinical and clinical advances in photodynamic
therapy. Photochem. Photobiol. 85 (2009) 1053–1074.
[
[
[
[
[
16]. G. Ulrich, R. Ziessel, A. Harriman, The chemistry of fluorescent bodipy dyes:
Versatility unsurpassed. Angew. Chem. Int. Ed. 47 (2008) 1184–1201.
17]. S.G. Awuah, Y. You, Boron dipyrromethene (BODIPY)-based photosensitizers for
photodynamic therapy. RSC Adv. 2 (2012) 11169–11183.
18]. A. Kamkaew, S.H. Lim, H.B. Lee, L.V. Kiew, L.Y. Chung, K. Burgess, BODIPY
dyes in photodynamic therapy. Chem. Soc. Rev. 42 (2013) 77–88.
19]. B.G. Davis, M.A. Robinson, Drug delivery systems based on sugar-macromolecule
conjugates. Curr. Opin. Drug Discov. Develop. 5 (2002) 279–288.
20]. X. Zheng, R.K. Pandey, Porphyrin-carbohydrate conjugates: impact of carbohydrate
moieties in photodynamic therapy (PDT). Anticancer Agents Med. Chem. 8 (2008)
2
41–268.