Tetra(3,4-pyrido)porphyrazines Caught in the Cationic Cage: Toward Nanomolar Active Photosensitizers

Article abstract of DOI:10.1021/acs.jmedchem.6b01140

Investigation of a series of tetra(3,4-pyrido)porphyrazines (TPyPzs) substituted with hydrophilic substituents revealed important structure-activity relationships for their use in photodynamic therapy (PDT). Among them, a cationic TPyPz derivative with total of 12 cationic charges above, below and in the plane of the core featured a unique spatial arrangement that caught the hydrophobic core in a cage, thereby protecting it fully from aggregation in water. This derivative exhibited exceptionally effective photodynamic activity on a number of tumor cell lines (HeLa, SK-MEL-28, A549, MCF-7) with effective concentrations (EC₅₀) typically below 5 nM, at least an order of magnitude better than the EC₅₀ values obtained for the clinically approved photosensitizers verteporfin, temoporfin, protoporphyrin IX, and trisulfonated hydroxyaluminum phthalocyanine. Its very low dark toxicity (TC₅₀ > 400 μM) and high ability to induce photodamage to endothelial cells (EA.hy926) without preincubation suggest the high potential of this cationic TPyPz derivative in vascular-targeted PDT.

Full text of DOI:10.1021/acs.jmedchem.6b01140

Subscriber access provided by HKU Libraries
Article
Tetra(3,4-pyrido)porphyrazines caught in the cationic
cage: toward nanomolar active photosensitizers
Miloslav Machacek, Jiri Demuth, Pavel Cermak, Magda Vavreckova, Lenka Hruba,
Adela Jedlickova, Pavel Kubat, Tomas Sim#nek, Veronika Novakova, and Petr Zimcik
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01140 • Publication Date (Web): 28 Sep 2016
Downloaded from http://pubs.acs.org on October 1, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Tetra(3,4ꢀpyrido)porphyrazines caught in the
cationic cage: toward nanomolar active
photosensitizers
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Miloslav Machacek,† Jiri Demuth,‡ Pavel Cermak,‡ Magda Vavreckova,§ Lenka Hruba,‡ Adela
Jedlickova,† Pavel Kubat,ǁ Tomas Simunek,† Veronika Novakova*,‡ and Petr Zimcik*,§
†
Department of Biochemical Sciences, Faculty of Pharmacy in Hradec Kralove, Charles
University in Prague, Heyrovskeho 1203, Hradec Kralove, 500 05, Czech Republic
‡
Department of Biophysics and Physical Chemistry, Faculty of Pharmacy in Hradec Kralove,
Charles University in Prague, Heyrovskeho 1203, Hradec Kralove, 500 05, Czech Republic
§
Department of Pharmaceutical Chemistry and Pharmaceutical Analysis, Faculty of Pharmacy in
Hradec Kralove, Charles University in Prague, Heyrovskeho 1203, Hradec Kralove, 500 05,
Czech Republic
ǁJ. Heyrovský Institute of Physical Chemistry, v.v.i., Academy of Sciences of the Czech
Republic, Dolejškova 3, 182 23 Praha 8, Czech Republic
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Investigation of a series of tetra(3,4ꢀpyrido)porphyrazines (TPyPzs) substituted with hydrophilic
substituents revealed important structureꢀactivity relationships for their use in photodynamic
therapy (PDT). Among them, a cationic TPyPz derivative with total of twelve cationic charges
above, below and in the plane of the core featured a unique spatial arrangement that caught the
hydrophobic core in a cage, thereby protecting it fully from aggregation in water. This derivative
exhibited exceptionally effective photodynamic activity on a number of tumor cell lines (HeLa,
SKꢀMELꢀ28, A549, MCFꢀ7) with effective concentrations (EC ) typically below 5 nM, at least
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
an order of magnitude better than the EC50 values obtained for the clinically approved
photosensitizers verteporfin, temoporfin, protoporphyrin IX and trisulfonated hydroxyaluminium
phthalocyanine. Its very low dark toxicity (TC50 > 400 ꢁM) and high ability to induce
photodamage to endothelial cells (EA.hy926) without preincubation suggest the high potential of
this cationic TPyPz derivative in vascularꢀtargeted PDT.
ACS Paragon Plus Environment
2

Page 3 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Photodynamic therapy (PDT) is a treatment modality of cancer that has been well established
1
in clinical practice since the approval of the first drug, porfimer sodium, in 1993. PDT combines
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
three essentially nonꢀtoxic components ꢀ a photosensitizer (PS), light and oxygen. The PS
absorbs the energy of the activating light and transfers it to groundꢀstate oxygen, with subsequent
2
formation of highly reactive singlet oxygen that destroys the target cells. Even though several
PSs have been clinically approved in the last twenty years, there is still room for improvement in
the field of PDT to limit the toxicity and improve the efficiency of the drugs and/or increase the
targeting ability of PSs to cancer cells. Several strategies have been employed to obtain more
efficient and less toxic PSs. Conjugation with tumorꢀspecific ligands (such as small molecules,
peptides, proteins and antibodies) substantially increases the selectivity of the active PS toward
3ꢀ5
specific cells. The systemic toxicity may also be reduced by PSs that circulate in their inactive
form in the body and are activated only by specific conditions in the tumor tissues (e.g., pH,
6
ꢀ8
reducing environment). Other less traditional strategies, such as the preparation of hollow PS
9
10
nanospheres or dual PSs, also improved the photodynamic effectiveness.
However, photodynamic treatment is already highly selective per se due the irradiation of
target tissues only. Therefore, the improved selectivity due to the specific localization of PS
1
1,12
appears to not be absolutely necessary unless diagnostic or theranostic effects are required.
An interesting strategy to decrease the toxicity by reducing the overall exposure to the drug
appears to involve rapid elimination of the strongly hydrophilic PSs from the body at the end of
1
3
the efficient treatment. This approach requires highly hydrophilic PSs with rather limited
1
4
protein binding as a prerequisite for rapid clearance.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Phthalocyanines (Pcs) belong to a promising family of PSs with high singlet oxygen
production, strong absorption over 680 nm and wide variability in substitution that may finely
1
5
tune the required photophysical and physicoꢀchemical properties. For example, trisulfonated
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
hydroxyaluminium Pc (S AlOHPc) has been clinically approved for PDT in Russia since 2001.
3
1
7,18
Additionally, the azaꢀanalogues of Pcs, tetrapyrazinoporphyrazines
and tetra(3,4ꢀ
1
9
pyrido)porphyrazines (TPyPzs) , are active PSs. This work focused on the synthesis and
thorough in vitro assessment of the photodynamic properties of a group of TPyPzs substituted
with hydrophilic substituents to reveal their potential in PDT in direct comparison with other
clinically used PSs. The studied TPyPzs 1ꢀ6 (Chart 1) were functionalized with various
hydrophilic moieties (anionic, cationic, or nonꢀcharged) that were attached to the core through O
or S bridges with flexible (1ꢀ5) or bulky and rigid (6) linkers. Nitrogen as the bridging
19
heteroatom appeared to be less suitable for PDT from a photophysical point of view. In
addition to the introduction of the hydrophilic moieties, the selection of the substituents allowed
assessment of the effect of specific functional groups on the properties of the TPyPzs. In
particular, we were interested in the inhibition of TPyPz aggregation by these substituents
2
0
because this property appears to be crucial for efficient photodynamic activity of Pcꢀlike PSs.
ACS Paragon Plus Environment
4

Page 5 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chart 1. Structures of the investigated TPyPzs. Only one isomer is drawn.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
a
Scheme 1. Synthesis of dicarbonitrile precursors and corresponding TPyPzs.
a
Reaction conditions: i) 2ꢀmercaptoethanol, NaOH, THF/H O, rt; ii) 3ꢀmercaptopropanoic acid,
2
K CO , DMSO, rt; iii) N,Nꢀdiethylaminoethanethiol hydrochloride, K CO , DMSO, rt; iv) N,Nꢀ
2
3
2
3
diethylaminoethanol, NaH, rt, Ar; v) 2,6ꢀbis((1Hꢀimidazolꢀ1ꢀyl)methyl)ꢀ4ꢀmethylphenol, NaH,
DMF, rt, Ar; vi) Mg, butanol, reflux, then TsOH, AcOH/H O, rt, then Zn(AcO) , pyridine,
reflux; vii) Mg, butanol, reflux, then TsOH, THF/MeOH, rt, then Zn(AcO) , pyridine, reflux,
2
2
2
then NaHCO ; viii) Mg, butanol, reflux, then TsOH, THF/MeOH, rt, then Zn(AcO) , pyridine,
3
2
reflux, then CH CH I, NMP, rt; ix) Mg, butanol, reflux, then TsOH, THF/MeOH, rt, then
3
2
Zn(AcO) , pyridine, reflux, then CH I, NMP, rt; x) Mg, butanol, reflux, then 1% HCl, rt, then
2
3
Zn(AcO) , pyridine, reflux, then CH CH I, NMP, rt; xi) Mg, butanol, reflux, then 1% HCl, rt,
2
3
2
then Zn(AcO) , pyridine, reflux, then CH I, DMF, 80°C.
2
3
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
1
Synthesis. Surprisingly, since their first synthesis by Linstead in 1937, the synthetic
modifications of TPyPzs have been limited to the exchange of central cations and/or the
quaternization of pyridine nitrogen. We only recently suggested a method for the simple
functionalization of TPyPzs starting from 2ꢀchloroꢀ5,6ꢀdimethylpyridineꢀ3,4ꢀdicarbonitrile (7)
and thus paved the way for their rational design and fine tuning of their properties for selected
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9
applications. Substituted pyridineꢀ3,4ꢀdicarbonitriles 8ꢀ12, precursors for cyclotetramerization,
were obtained from 7 by nucleophilic substitution in reasonable yields that were typically over
7
0% (Scheme 1). Only in the case of 11 was the yield lower, at 35%, due to the number of side
products. Linstead cyclotetramerization using magnesium butoxide as the initiator was employed
to obtain MgTPyPzs (mostly as a mixture of four positional isomers) because the direct template
19
method did not produce the TPyPz ring as reported previously. The metalꢀfree ligands were
released from the Mg complexes by treatment with acids and converted to Zn complexes with
zinc acetate in pyridine. Compounds 3ꢀ6 contained quaternized peripheral moieties that
originated from reaction with alkyl iodide. Significant differences in reactivity were observed
during alkylation of Oꢀ and Sꢀsubstituted TPyPzs. A strong electronꢀdonating effect of alkyloxy
and aryloxy substituents increased the electron density on the pyridine nitrogen of the core and
consequently led to its quaternization (compounds 5 and 6). Alkylation at both the pyridine and
imidazolyl nitrogens was confirmed by signals at 54.6 ppm and 36.0 ppm, respectively, in the
13
C NMR of 6. The pyridine nitrogen was not alkylated in TPyPzs bearing an S bridge (3 and 4).
All TPyPzs were primarily characterized by elemental analysis and MALDIꢀTOF
spectrometry. In addition to the presence of a mixture of positional isomers, the characterization
of the final TPyPzs by NMR was further complicated by their tendency to aggregate, which led
ACS Paragon Plus Environment
6

Page 7 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to broad and flat signals that could not be used to clearly identify 1, 2 or 5 by NMR technique.
This aggregation tendency is, however, very typical for TPyPz macrocycles, and very few papers
1
22,23
have reported H NMR data for this kind of compounds.
Broad but much less flat proton
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
signals were obtained for 3 and 4, supporting their structure. The bulkiness and rigidity of the
substituted phenoxy groups resulted in sharp proton signals for both 6 and its nonꢀalkylated
macrocycle, suggesting that aggregation can be avoided in such a way. Detailed analysis of the
1
H NMR spectra revealed the presence of two major positional isomers C and C in a ratio of
4
h
2v
approximately 3:1, although a minor presence of other isomers (C and D ) cannot be
s
2h
unequivocally excluded. The prevalence of the least sterically congested isomer in the mixture is
understandable considering that the extremely bulky substituted phenoxy groups cannot be easily
1
3
oriented faceꢀtoꢀface in adjacent units. Interestingly, these isomers were not observed in the C
NMR spectrum (only one signal for each carbon was typically observed).
Absorption spectra. The absorption spectra of TPyPzs 1 and 3ꢀ6 in DMF contained the
typical features of TPyPz spectra and corresponded to the prevalently monomeric form (Fig. 1).
Only the absorption spectra of compound 2 indicated extensive aggregation, even in DMF. The
position of the main absorption Qꢀband was influenced by the type of heteroatom bridge.
Compounds 1ꢀ4 with sulphur as the linker typically absorbed at approximately 720 nm, whereas
the Oꢀisosteres were slightly blueꢀshifted by approximately 10 nm (Table 1). The shift of the Qꢀ
2
4,25
band agrees with the typical behavior of Oꢀ and Sꢀsubstituted Pcs and their azaꢀanalogues.
In
comparison with the clinically used PSs (Table 1), TPyPzs 1ꢀ6 absorbed light of longer
wavelengths, which is advantageous in PDT because of the deeper penetration of the excited
2
light into the tissues.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The planar hydrophobic core of Pcs and their analogues is typically prone to extensive selfꢀ
1
8,26,27
aggregation in water.
Only a few rare examples of Pcs truly nonꢀaggregating in water have
1
3,20,28ꢀ31
been reported in the literature.
When dissolved in cell culture medium (full medium
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
containing 10% fetal bovine serum (FBS)), TPyPzs 1ꢀ5 were characterized by a broad Qꢀband
with a decreased extinction coefficient and the appearance of new blueꢀshifted bands (Fig. 1).
These features indicated that the hydrophilic substituents on flexible linkers did not protect the
hydrophobic cores from selfꢀaggregation. In contrast, the rigid arrangement of quaternized
imidazoles in 6, which were forced into positions above and below the TPyPz plane, perfectly
protected the macrocycle molecules from aggregating together due to electrostatic repulsive
forces. Such an arrangement has been recently shown to be very efficient in hexadecacationic Pc
2
0
and resulted in high photodynamic activity. The advantage of TPyPz 6 lies in the additional
cationic charges placed in the plane of the core on quaternized pyridines. In this way, the
hydrophobic core is caught in the cage by the cationic charges and is completely protected from
above, below and the side. As a consequence of this unique structural feature, no aggregation
was detected in cell culture medium, even at a concentration of 500 ꢁM (Fig. S24). The
monomeric character of 6 was affected (Fig. S25) neither by changes in the ionic strength (water,
buffer), by pH (from 3 to 9) nor by the presence of proteins (cell culture media).
ACS Paragon Plus Environment
8

Page 9 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Absorption spectra of the studied TPyPzs 1 (a), 2 (b), 3 (c), 4 (d), 5 (e) and 6 (f) in
DMF (red) and cell culture medium (blue) at a concentration of 1 ꢁM. The samples 1ꢀ5 in cell
culture medium were prepared from a 100 ꢁM DMF stock solution.
Photophysical characterization. The photophysical parameters that are important for
photodynamic treatment were determined primarily in DMF (Table 1). TPyPzs 1ꢀ6 emitted red
fluorescence upon excitation, and their emission spectra were mirror images of their absorption
spectra, with small Stokes shifts not exceeding 15 nm. The excitation spectra can be used for fine
assessment of aggregation. They perfectly matched the absorption spectra, confirming the fully
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
monomeric character of 3, 4 and 6 in DMF (Fig. S16). The slight increase in absorption at
approximately 670 nm for 1 and 5 indicated a very minor contribution from aggregates that was
also reflected in the lower quantum yields of these compounds (see discussion below).
Compound 2 fully aggregated in DMF, as suggested above. The fluorescence quantum yields
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(ΦF) were typically in the range 0.11 – 0.13 for nonꢀaggregated 3, 4 and 6 and were 0.09 for the
slightly aggregated TPyPzs; very weak fluorescence was detected for TPyPz 2. The fluorescence
lifetimes were not greatly affected by aggregation (the signal is due only to the fluorescent
species), which explains the τF values between 1.5 and 2 ns for all TPyPzs, including 2.
Considering the singlet oxygen production, all TPyPzs were excellent sensitizers, with high
singlet oxygen quantum yields (Φꢀ) that reached values of approximately 0.70 for nonꢀ
aggregated 3, 4 and 6. Similar to ΦF, the small aggregation of 1 and 5 in DMF lowered Φꢀ to ~
0.42. The low Φꢀ ~ 0.09 of 2 is due to extensive aggregation in DMF. The very small transient
absorption signal of the triplet states of 2, together with their efficient quenching by oxygen,
indicated poor photogeneration of the triplet states but efficient production of singlet oxygen
from these states. The lifetime of singlet oxygen (τꢂ) in DMF varied from 9.0 to 19.3 ꢁs,
1
illustrating the high ability of inherent impurities to quench O ( ꢀ ), which is typical for DMF as
2
g
32
the solvent.
Transient absorption measurements revealed the effect of the bulky peripheral substituents in 6
that partially prevented the diffusion of oxygen to the chromophore. This effect was documented
by a lower rate constant k (Fig. S18) or a slightly lower fraction of the triplet states trapped by
O2
oxygen in 6 in an airꢀsaturated solution (Table 1) compared to those of all other tested
compounds with a relatively planar character and with no limitation of oxygen access to the core.
ACS Paragon Plus Environment
10

Page 11 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Although a similar effect was also observable for 6 in D O, it did not substantially affect the
2
efficient production of singlet oxygen by this PS.
Water typically represents a problematic medium for various PSs due to their extensive
aggregation or limited solubility (e.g., verteporfin, temoporfin and protoporphyrin IX (Pp IX) are
not water soluble without a coꢀsolvent). The ΦF values of 6 in water were the same as in DMF,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
and only a slight decrease of Φꢀ to 0.58 in D O was observed, suggesting that 6 retains excellent
2
photosensitizing properties even in water.
The photophysical characterization was also performed for clinically used PSs (S AlOHPc,
3
verteporfin, temoporfin, Pp IX) and methylene blue (MB) to serve as reference compounds in the
following biological evaluations (Table 1).
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
a
Table 1. Absorption and photophysical data of the studied compounds.
b
ꢀ9
O2
b
Cpd.
Solvent λmax
log ε λF
(nm)
τF (ns)
ΦF
τT (ꢁs)
oxygen air argon
10 k
FT
Φꢀ
τꢀ (ꢁs)
O2
1 ꢀ1
(
nm)
ꢀ
(l.mol s )
3.3±0.0
4.0±0.1
3.2±0.0
3.1±0.1
3.4±0.0
1.1±0.0
4.4±0.4
2.7±0.0
3.9±0.1
3.9±0.1
3.8±0.1
1
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
729
718
722
722
711
710
669
681
689
630
651
5.16
4.49
5.41
5.33
4.88
5.28
ꢀ
744
736
734
734
722
721
674
687
694
633
655
1.75
1.89
1.51
1.56
1.92
1.97
0.09
0.01
0.11
0.10
0.09
0.13
0.28
0.57
0.14
0.04
0.096
0.081
0.102
0.100
0.094
0.29
0.45 131
0.42 230
0.50 124
0.53 127
0.45 114
1.38 89
0.997 0.42
15.8
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
2
0.998 ~0.09 9.0
3
0.996 0.72
0.996 0.72
0.999 0.43
0.984 0.69
0.998 0.56
1.000 0.37
0.999 0.87
0.989 0.13
0.999 0.55
14.7
4
14.3
9.4
5
6
9.5
c
c
c
c
ZnPc
3.30
0.072
0.120
0.061
0.081
0.083
0.31 212
0.57 1630
0.29 370
0.42 37
19.3
19.2
18.9
19.1
18.6
S3AlOHPc
verteporfin
Pp IX
5.25
4.57
3.87
4.50
5.08
6.12
7.61
9.29
temoporfin
0.20
0.42 214 (69 %)
1
050 (31 %)
d
6
H
2
O
711
665
5.21
4.82
724
685
2.19
0.13
0.81
0.36
2.66 64
0.84±0.08
2.0±0.1
0.958 0.58
0.986 0.52
61.3
61.1
d
MB
H O
0.40 (86%) 0.04
.99 (14%)
5.98
1.62 113
2
4
d
S3AlOHPc
H O
679
5.17
684
0.46
0.42
1.86 833
1.7±0.1
0.999 0.37
55.3
2
a
Qꢀband absorption maximum (λmax); extinction coefficient (ε); emission maximum (λ ); fluorescence lifetime (τ ); fluorescence
quantum yield (ΦF); triplet state lifetime (τ ); rate constant for quenching of triplet states by oxygen (k ); fraction of triplet states
deactivated by oxygen (FT ); singlet oxygen quantum yield (Φꢀ); lifetime of singlet oxygen (τ ). The following reference
compounds and values were used: MB (Φꢀ = 0.52 in D O ), ZnPc (Φꢀ = 0.56 in DMF , ΦF = 0.32 in THF ). Data from ref. . The
F
F
T
O2
O2
b
ꢀ
3
3
34
24
c
24 d
2
O2
following parameters were determined in D O: τ , k , F
, Φꢀ, τꢀ.
2
T
O2
T
ACS Paragon Plus Environment
1
2

Page 13 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vitro evaluation of the dark toxicity and phototoxicity. Comparison of the photodynamic
activities of various PSs reported in the literature is typically complicated by different testing
conditions, such as the irradiation (nature and energy of the light source, excitation wavelengths,
bandwidth, etc.), incubation time, cell line, viability assay used, and different presentations of the
results. For this reason, we evaluated a set of four clinically established PSs and MB (which is
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
5,36
not approved but was widely studied in the past as a potential PS
) in vitro and directly
compared their photodynamic activities under identical conditions. S AlOHPc is sulfonated
3
hydroxyaluminium Pc (average degree of sulfonation, n = 3) and has been approved in Russia
16
37
38
since 2001. Temoporfin and verteporfin are examples of the chlorin structural family with
more than fifteen years of worldwide clinical use. Pp IX, with a porphyrin macrocycle, is not
used directly as a PS in PDT but represents an active form of a widely used prodrug δꢀ
39
aminolevulinic acid and its esters.
To demonstrate the potential of all the studied compounds in PDT, the dark toxicity (expressed
as the half maximal toxic concentration, TC ) and photodynamic activity (expressed as the half
5
0
2
maximal effective concentration, EC ) after irradiation (λ > 570 nm, 11.2 J/cm ) were initially
5
0
assessed on a human cervical carcinoma cell line (HeLa). Compounds 1 and 2 proved to be
photodynamically ineffective under these experimental conditions due to the extensive
aggregation in the cell culture medium, as stated above. Compounds 3-5 displayed high
photodynamic activity (Table 2, Fig. 2a) that was comparable with several recently published
18
Pcs and their analogues, and compound 6 induced an exceptionally high photodynamic effect
with EC = 3.8 ± 0.2 nM. The clinically established PSs listed above proved to be significantly
50
less effective than 6 under the same experimental conditions (Table 2, Fig. S28). Verteporfin and
temoporfin were less active by an order of magnitude, whereas almost three orders of magnitude
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
higher EC values were determined for MB, Pp IX and S AlOHPc, making compound 6 one of
5
0
3
the most active PS reported to date, to the best of our knowledge. For a comparison with the
literature data on highly active phthalocyanine photosensitizers tested under similar conditions,
see Table S1.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
An effort was made to correlate the EC50 data of all studied compounds with their
lipophilicities expressed as log P (Table 2). Clear increase of activity with decreased lipophilicity
was observed for TPyPzs (Fig. 3a). However, this cannot be considered as the general rule for
PSs since rather reverse relationship was apparent for clinically approved PSs where the most
active temoporfin and verteporfin were characterized by the highest log P (Fig. 3a). As the
singlet oxygen is supposed to be the most important cytotoxic species, the ability to produce
singlet oxygen was correlated with the photodynamic activity. Expectedly, the compounds with
the highest singlet oxygen production were also the most phototoxic (e.g. verteporfin,
temoporfin, cpds. 3, 4, 6, Fig. 3b).
−
2
−2
Figure 2. a) Phototoxicity (λ > 570 nm, 12.4 mW cm , 15 min, 11.2 J cm ) and b) dark toxicity
against the HeLa cells of 4 (red), 5 (black), and 6 (blue). Five independent experiments, each in
quadruplicate, were typically performed.
ACS Paragon Plus Environment
14

Page 15 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Dependence of photodynamic activity of studied TPyPzs (red dots) and approved PSs
(
blue squares) on their lipophilicity (a) or ability to produce singlet oxygen in DMF (in D O for
2
MB) (b).
The dark toxicity, i.e., the inherent toxicity of the compounds without irradiation, could not be
determined for 1, 2 and Pp IX due to the relatively low limit of solubility. The TC values of the
5
0
other TPyPzs were over 100 ꢁM (Fig. 2b), indicating negligible toxicity that was several times
better than the toxicities of the established PSs with typical TC values of approximately 20 ꢁM
5
0
(
Table 2). The latter TC values correspond well with those determined also by other groups
50
5
(e.g. TC50 = 16.1 ꢁM for temoporfin on PCꢀ3M cells). Only S AlOHPc had TC values
3
50
comparable to those of the TPyPzs. Notably, the lowest toxicity was observed for 6, which,
together with its excellent photodynamic activity, resulted in an exceptionally favourable
“
therapeutic ratio” TC /EC of over 110 000. This value exceeded the TC /EC of any other
50 50 50 50
compound in this study by at least two orders of magnitude, making TPyPz 6 a very promising
PS for the photodynamic treatment of cancerous diseases.
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Comparison of the log P, dark toxicity (TC values) and photodynamic activity (EC
50
50
a
values) of the studied PSs on HeLa cells.
Cpd.
log P EC (μM)
TC (μM) TC /EC
50
5
0
50
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
b
1
0.90
> 50
> 50
> 50
> 50
ꢀ
ꢀ
c
b
b
2
ꢀ
d
d
3
ꢀ0.71 0.26 ± 0.089
ꢀ0.81 0.15 ± 0.041
105 ± 9.5
154 ± 4.5
115 ± 46
400
4
1 000
130
5
0.91
0.87 ± 0.068
6
ꢀ1.99 0.0038 ± 0.0002 435 ± 26
0.65 5.15 ± 1.93 22.7 ± 7.9
110 000
4.4
MB
S3AlOHPc ꢀ2.12 2.07 ±0.29
127 ± 7.8
18.6 ± 4.2
60
verteporfin 1.56
Pp IX 0.06
0.036 ± 0.010
0.80 ± 0.13
510
b
> 10
> 10
460
temoporfin 1.15
0.045 ± 0.007
20.6 ± 2.5
a
Data are presented as the TC or EC values ± standard deviations. Irradiation conditions: λ
5
0
50
−
2
−2 b
> 570 nm, 12.4 mW cm , 15 min, 11.2 J cm . The compounds precipitated at concentrations
over the listed values. Could not be determined due to the precipitation. Data from ref.
c
d
19
Due to the short lifetime of singlet oxygen, the PSs only affect the biological structures in their
imminent vicinity – thus, subcellular localization is crucial to understanding the processes
leading to cell death upon photodynamic treatment. To evaluate the subcellular localization of 1-
6
, specific fluorescent probes for mitochondria and lysosomes were employed along with the
intrinsic fluorescence of the TPyPzs. All of the studied TPyPzs were located exclusively in the
endoꢀlysosomal compartment because they were detected as punctate fluorescence coꢀlocalized
with the lysosomal probe (Fig. S29). For 6, this localization was also confirmed using liveꢀcell
laser scanning confocal microscopy (Fig. 4).
ACS Paragon Plus Environment
16

Page 17 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Subcellular localization of 6 (red, B) in HeLa cells visualized by confocal microscopy
after coꢀincubation with the organelleꢀspecific fluorescent probes MitoTracker (green, D) and
LysoTracker (blue, C). A merged image is shown in Figure A. The lower figures show profiles
of the fluorescence intensity of the corresponding parts of the confocal image in part A.
Detailed in vitro studies of TPyPz 6. Based on its photophysical data, low inherent toxicity
and exceptional photodynamic activity, 6 was studied in more detail as the most promising
candidate for PDT. Additional cytotoxicity experiments were performed on several cell lines
(Fig. S28). The dark toxicity on nonꢀmalignant mouse fibroblasts (3T3) was found to be
similarly very low (TC = 427 ± 54 ꢁM). The photodynamic activity against different tumour
50
cell lines was established using human melanoma (SKꢀMELꢀ28), human lung carcinoma (A549),
human breast carcinoma (MCFꢀ7). Here, the photodynamic effects were similar to that in HeLa
cells for all malignant cell lines, reaching EC values of 3.77 ± 0.59 nM, 5.15 ± 1.57 nM and
5
0
2.83 ± 0.08 nM for SKꢀMELꢀ28, A549 and MCFꢀ7, respectively. Much lower phototoxicity was
induced on nonꢀmalignant cells, 3T3 fibroblasts and immortalised human HaCaT keratinocytes
with EC = 14.9 ± 5.30 nM and 42.3 ± 6.9 nM, respectively, suggesting a certain level of
50
selectivity for cancerous cells over healthy ones.High photodynamic effect (EC = 15.3 ± 1.8
5
0
nM, vide infra) was also observed against healthy permanent human umbilical vein endothelial
cell line (EA.hy926) suggesting that the selectivity is not a general rule and may be important for
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the selected cell lines only, e.g. in reducing potential skin phototoxicity due to lower effect
against keratinocytes.
Interestingly, high photodynamic activity was attained by 6 at a very low intracellular
concentration. The cellular uptake of 6 by HeLa cells was very rapid during the first two hours
and then continued steadily for the remainder of the 12 h incubation period (Fig. 5a). The total
amount of 6 after 12 h of incubation was 0.08 nmol/mg of protein in the cell lysate. When
comparison was possible with other compounds studied under the same conditions, the amount
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
0
of 6 was approximately seven or three times lower than the uptake of hexadecacationic Pc or
17
the hexadecaanionic azaꢀPc analogue , respectively, that both exerted much lower
photodynamic effect.
In addition to the singlet oxygen produced very efficiently even in water (Table 1), other
reactive oxygen species (ROS) also contributed to the lethal effect of 6 in cells. The intracellular
4
0
production of free radicals (HO•, RO•, NO • etc.) was monitored by their quencher,
2
chloromethylꢀmodified 2′,7′ꢀdichlorodihydrofluorescein diacetate (CMꢀH DCFDA), which is
2
41
reportedly insensitive to singlet oxygen. As observed from Fig. 5b, the free radicals were
produced only during irradiation of the cells, with negligible production of ROS before and after
activation of the PS by light.
ACS Paragon Plus Environment
18

Page 19 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 5. a) Cellular uptake of compound 6 by HeLa cells after incubation with 4 ꢁM 6. The
experiments were performed in duplicate. b) Increase in the fluorescence of 5(6)ꢀ
chloromethylꢀ2′,7′ꢀdichlorofluorescein (λ = 485 nm, λ = 525 nm) after the photodynamic
ex
em
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
treatment of HeLa cells incubated with
6
. The results were corrected for autoxidation by
subtracting the light and dark controls. The experiments were performed in triplicate.
The activity of 6 could be further increased by using a longer irradiation time (Fig. 6a).
Conversely, 6 was ineffective at killing cells when using a very short irradiation time (5 min)
under our experimental conditions probably due to its very low concentration in the cellular
compartments and the reversibility of the process at a low light dose.
Figure 6. a) Viability of HeLa cells as a function of the light dose following photodynamic
treatment with 6 (3.8 nM); green open bar, control cells (without 6 in the presence of light); blue
−
2
open bar, 5 min of irradiation (total fluence 3.7 J cm ); blue dashed bar, 15 min of irradiation
−2
−2
(
11.2 J cm ); blue full bar, 30 min of irradiation (22.3 J cm ). The experiments were performed
in duplicate. b) Flow cytometry assessment of the HeLa cell death mode for 6 analysed at
different times after irradiation at concentrations corresponding to the EC15 and the EC85,
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
expressed as the cell count based on the annexin V−Alexa Fluor 488 or/and PI positivity. Nonꢀ
apoptotic cells (green), apoptotic cells (orange), late apoptosis/early necrosis (red). The
experiments were performed in triplicate. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Indeed, cells displayed a mild morphological response to oxidative insult after 5 min of
irradiation. These changes were not irreversible or lethal at this length of time, and cells are
probably able to recover from this damage. A longer irradiation time (over 15 min) resulted in
massive cellular damage with distinct morphological features (see also Fig. 7). Cellular retraction
occurred shortly after the start of irradiation (5 min) and was followed by the formation of
42
massive membrane protrusions devoid of organelles (necrotic membrane blebs). Mitochondria
lost their typical filamentous structure and became spherical within 15 min of irradiation (Fig. 7).
Severely damaged cell membranes became permeable to propidium iodide (PI), which passed
into the cytoplasm and started to stain nuclei during the irradiation. In the postꢀirradiation period,
nuclei shrinkage and further PI staining of chromatin were observed. We were unable to detect
compound 6 by microscopic methods due to its very low intracellular concentration in this
experiment until its final relocalization to the nucleus (a result of the interaction of a cationic dye
with negatively charged DNA), which started 15 min after the irradiation period and ended
within one hour (Fig. 7). However, considering its multicationic character and rapid damage to
other organelles, we suggest that 6 was released from the lysosomes and relocalized into the
other subcellular compartments within a few minutes after activation, even when it could not be
easily visualized at the low concentration. All these changes observed during irradiation indicate
a fast photodynamic effect of 6. The morphological changes were corroborated by monitoring
the impairment of the membranes and cytoskeleton using confocal imaging of live and fixed
ACS Paragon Plus Environment
20

Page 21 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
specimens. The photodynamic action of 6 (at EC = 7.58 nM) resulted in complex changes –
85
pyknosis (chromatin condensation), large membrane protrusions and reorganization of the actin
and tubulin cytoskeleton (for more details see Figs. S29ꢀS31 with corresponding discussion in
the Supporting Information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These rapid and severe morphological changes, together with the PI staining of cell nuclei that
started during the irradiation period, suggest that necrosis was the main pathway of cell demise
as a result of lysosomal rupture and continuous severe oxidative damage to other cellular
components. Predominantly necrotic cell death has been further confirmed by flow cytometry
using coꢀstaining of the cells with PI and Annexin VꢀAlexa Fluor 488 – nearly no apoptotic cells
were detected using both low (EC ) and high (EC ) doses of 6 (Fig. 6b).
15
85
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. Changes in the morphology of HeLa cells treated with 6 during and after irradiation.
The nuclei were visualized with Hoechst 33342 (Hoechst) and propidium iodide (PI); the
mitochondria were visualized by MitoTracker Green. The arrows indicate important details
discussed in the text. DIC = differential interference contrast. Bar represents 200 ꢁm.
ACS Paragon Plus Environment
22

Page 23 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Considering that the highly hydrophilic and nonꢀaggregating character of 6 resulted in
excellent photodynamic activity at a low dose, we further assessed the potential of this
compound in vascularꢀtargeted PDT (VTP). VTP is a strategy that does not require any lag time
between PS administration and irradiation and damages mainly endothelial cells, leading to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
3
tumour death primarily by vascular shutdown. To evaluate whether 6 would be suitable for
VTP, its phototoxicity was assessed in a permanent human umbilical vein endothelial cell line
(
EA.hy926) without any preincubation. In this experimental setꢀup, 6 still proved to be highly
effective, with EC = 15.3 ± 1.8 nM. Additionally, 6 did not significantly bind to the serum
5
0
proteins, as deduced from a constant fluorescence intensity and Qꢀband position after the
addition of FBS to a solution of 6 in serumꢀfree medium (Fig. S26). This lack of binding is a
good prerequisite for rapid excretion from the body by glomerular filtration, which may limit
14
systemic toxicity.
Highlyꢀhydrophilic compounds could be administered intravenously directly without need of
any vehicle. In this case, it is crucial that drugs administered by this route do not induce any
44
undesirable hematological effect such as drugꢀinduced hemolysis. Thus, ex vivo red blood cell
hemolysis assay is highly recommended to determine the hemolytic potential of compounds
intended for intravenous administration. Experiments with washed and diluted human
erythrocytes incubated with the range of high concentration of 6 (10 – 1000 ꢁM) were
performed. No hemolysis was observed even at highest tested concentration (Fig. S27).
CONCLUSION
TPyPzs are a rather old structural group of macrocyclic dyes for which the potential in PDT
has been neglected for a long time. In this study, we demonstrated that suitable structural
modifications can dramatically increase the therapeutic potential of these dyes. Due to their
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
planar structure, TPyPzs suffer from aggregation, which is common for structurally related and
more commonly studied Pcs. The aggregation of TPyPzs in waterꢀbased media cannot be easily
solved using nonionic (1) or anionic substituents (2) which was also reflected in the lack of
photodynamic activity. The tendency to aggregation was improved with cationic TPyPzs with
flexible linkers (3ꢀ5). The EC values of these TPyPzs were only slightly higher than those of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
the best clinically approved PSs (verteporfin, temoporfin) and better than some of the other PSs
Pp IX, MB, S AlOHPc). All TPyPzs were characterized by lower inherent toxicity in the dark.
(
3
Additionally, direct comparison of the photodynamic properties of the clinically approved PSs
and MB) revealed that the activity against HeLa cells under our experimental conditions
(
decreased in the following order: verteporfin ~ temoporfin > Pp IX > S AlOHPc > MB.
3
A dramatic shift in the properties occurred when the TPyPzs were substituted nonꢀperipherally
with bulky cationic substituents, when the charges are forced above and below the macrocycle
core. This arrangement is supported by additional charges in the plane of the core from the
quaternized pyridine (a modification impossible for Pcs) that catch the hydrophobic TPyPz core
in the cationic cage and efficiently protect the molecules by electrostatic repulsive forces from
any aggregation. As a consequence, excellent photosensitizing properties of 6 were retained in
waterꢀbased media without the addition of any coꢀsolvents or surfactants and were reflected in
the outstanding photodynamic activity of 6 on several different tumour cell lines with EC50
values typically below 5 nM. This value is an order of magnitude lower than that observed for
the clinically approved PSs. Additionally, 6 was characterized by the lowest inherent toxicity
(
TC = 435 ± 26 ꢁM on HeLa cells), which was at least an order of magnitude better than the
50
toxicity of any established PS. This hydrophilic TPyPz also has potential in VTP because it
efficiently destroys endothelial cells upon irradiation, even without preincubation. On the
ACS Paragon Plus Environment
24

Page 25 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
subcellular level, this PS relocalized from lysosomes (primary target) to other organelles and
eventually to the nucleus after irradiation. Impairment of the subcellular structures led to a
predominantly necrosisꢀtype cell death.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In conclusion, the versatility of the TPyPz structure enabled the development of PS 6, which is
more efficient and less toxic than any of the clinically approved PSs for PDT.
EXPERIMENTAL SECTION
General Methods. All of the organic solvents used in the synthesis were of analytical grade.
Anhydrous butanol for the cyclotetramerization was freshly distilled from magnesium.
Unsubstituted zinc phthalocyanine (ZnPc) was purchased from SigmaꢀAldrich. All of the other
chemicals for the syntheses were purchased from certified suppliers (i.e., SigmaꢀAldrich, TCI
Europe, Acros, and Merck) and used as received. Thin layer chromatography was performed on
Merck aluminum sheets coated with silica gel 60 F254. Merck Kieselgel 60 (0.040−0.063 mm)
was used for column chromatography. The melting points were measured on an Electrothermal
IA9200ꢀseries digital meltingꢀpoint apparatus (Electrothermal Engineering, SouthendꢀonꢀSea,
Essex, Great Britain). The infrared spectra were measured on a Nicolet 6700 spectrometer in
1
13
ATR mode. The H and C NMR spectra were recorded on a VNMR S500 NMR spectrometer.
The chemical shifts are reported relative to Si(CH ) and were locked to the signal of the solvent.
3
4
J values are given in Hz. The UV−Vis spectra were recorded using a Shimadzu UVꢀ2600
spectrophotometer. Elemental analysis was carried out using a Vario Micro Cube Elemental
Analyzer (Elementar Analysensysteme GmbH, Hanau, Germany). The MALDIꢀTOF mass
spectra were recorded in positive reflectron mode on a 4800 MALDI TOF/TOF mass
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
spectrometer (AB Sciex, Framingham, MA, USA) in transꢀ2ꢀ[3ꢀ(4ꢀtertꢀbutylphenyl)ꢀ2ꢀmethylꢀ2ꢀ
propenylidene]ꢀmalononitrile, which was used as a matrix. The instrument was calibrated
externally with a fiveꢀpoint calibration using a Peptide Calibration Mix1 kit (LaserBio
Laboratories, Sophiaꢀ Antipolis, France). Starting material 2ꢀchloroꢀ5,6ꢀdimethylpyridineꢀ3,4ꢀ
dicarbonitrile (7), 2ꢀ(2ꢀ(diethylamino) ethylsulfanyl)ꢀ5,6ꢀdimethylpyridineꢀ3,4ꢀdicarbonitrile
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(10), TPyPz 3 and its nonꢀalkylated macrocyclic precursor (3a) were prepared according to
1
9
literature. All of the compounds gave satisfactory elemental analyses with a difference of ≤
0.4% from the calculated values indicating ≥95% purity.
Preparation of 1,8,15,22-tetrakis(2-hydroxyethylsulfanyl)-3,4,10,11,17,18,24,25-
octamethyltetra(3,4-pyrido)porphyrazinato zinc(II) (1). Magnesium (729 mg, 30 mmol) and a
small crystal of iodine were refluxed in freshly distilled anhydrous butanol (10 mL) until all Mg
was converted to magnesium butoxide (approx. 4 h). Compound 8 (989 mg, 4.2 mmol) was
added, and the reflux continued for another 15 h. After cooling, the solvent was evaporated, and
the product was dissolved in 50% aq. AcOH, filtered and evaporated to dryness. The dark solid
was washed thoroughly with hot water, acetone and THF. Then, the product was dissolved in a
small amount of pyridine and precipitated by addition of acetone. The precipitate was collected
and purified on silica eluting first with MeOH to elute impurities and subsequently with
pyridine:ethyl acetate 3:1 to yield 167 mg (16%) of a green solid that was characterized by
+
+
MALDIꢀTOF (m/z 957.2 [M + H] ; calculated for C H MgN O S + H : 957.2). The Mg
4
4
44
12
4 4
complex was directly converted to the metalꢀfree ligand upon dissolution in 80% aq. AcOH (30
mL) with addition of pꢀtoluenesulfonic acid hydrate (TsOH, 323 mg, 1.7 mmol). The mixture
was stirred at rt for 4 h, the solvents were evaporated, and the dark solid was washed thoroughly
+
with water and acetone to give a dark solid of the metalꢀfree ligand (m/z 935.2 [M + H] , 917.2
ACS Paragon Plus Environment
26

Page 27 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
+
[
M ꢀ OH] ; calculated for C H N O S + H : 935.3). The whole product was subsequently
44
46 12
4 4
dissolved in pyridine (10 mL) and refluxed for 4 h with zinc acetate (329 mg, 1.8 mmol). The
pyridine was evaporated, and the product was washed with water and acetone. The dark solid
was subsequently adsorbed onto silica, and the impurities were eluted with MeOH. The product
was eluted from the silica with pyridine:ethylacetate 3:1 and evaporated to dryness to yield a
dark green solid (90 mg, 9% based on 8). Anal. Calcd. for C H N O S Zn + 2H O: C, 51.08;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
4
44 12
4
4
2
3
ꢀ
H, 4.68; N, 16.25; Found: C, 51.46; H, 4.52; N, 15.96. λmax (pyridine, 1 µM)/nm 739 (ε/dm mol
1
ꢀ1
3
ꢀ1
ꢀ1
cm 170 200), 663 (45 100), 365 (48 000). λ (DMF)/nm 729 (ε/dm mol cm 144 600),
max
6
57 (43 500), 360 (43 800). IR (ATR) ν = 2922 (CH), 2855 (CH), 1553, 1472, 1406, 1266, 1229,
ꢀ1
1
185, 1109, 1023 cm . Due to aggregation, NMR spectra gave only broad and flat signals that
+
could not be used for structure evaluation. m/z (MALDI TOF) 997.1 [M + H] , calculated for
+
C H N O S Zn + H : 997.2.
44
44 12
4 4
Preparation of tetrasodium salt of 1,8,15,22-tetrakis(2-carboxyethylsulfanyl)-
3
,4,10,11,17,18,24,25-octamethyltetra(3,4-pyrido)porphyrazinato zinc(II) (2). Magnesium
(
618 mg, 25 mmol) and a small crystal of iodine were refluxed in freshly distilled anhydrous
butanol (30 mL) until all Mg was converted to magnesium butoxide (approx. 4 h). Compound 9
950 mg, 3.6 mmol) was added and the reflux continued for another 4 h. After cooling, the
(
solvent was evaporated, and 15% aq. AcOH was added (100 mL), and the mixture was stirred for
30 min at rt. The dark precipitate was collected and washed with water. The Mg complex was
directly converted to metalꢀfree ligand upon dissolution in THF:MeOH (1:1, 50 mL) with
addition of TsOH (1.2 g, 6.3 mmol). The mixture was stirred at rt for 30 min, the solvents were
evaporated, and the dark solid washed thoroughly with water. The whole product was
subsequently dissolved in pyridine (20 mL) and refluxed for 1 h with zinc acetate (2.5 g, 13.6
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mmol). The pyridine was evaporated and the product was washed with water. The solid was
suspended in a small amount of THF, 10% HCl was added (100 mL) and the suspension was
filtered and washed with 2% HCl and water. The product was purified by column
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
chromatography on silica with step gradient first eluting impurities with CHCl :MeOH:AcOH
3
10:1:1 with subsequent elution of the product with CHCl :MeOH:AcOH 5:2:1. The product was
3
dissolved in a mixture of 15% aq. NaOH (50 mL) and MeOH (2 mL), sonicated and filtered. The
filtrate was acidified with HCl and the fine precipitate was collected and washed with water.
Subsequently, the solid was dissolved in a small amount of DMF, precipitated by addition of
acetone, the precipitate was collected and washed with acetone, MeOH, CHCl and hexane to
3
yield a dark green solid (300 mg, 46% based on 9). The free acid was converted to its tetra
sodium salt upon dissolution in 4.5 equivalents of 1 M aq. NaOH and the salt was purified by gel
filtration on Sephadex Gꢀ25 eluting with water. Anal. Calcd. for C H N O S Zn (free acid) +
48
44 12
8 4
3
H O: C, 49.50; H, 4.33; N, 14.43%; Found (for free acid): C, 49.83; H, 4.25; N, 14.63; No
2
signal was detected in NMR spectra. No signal was detected in MALDI TOF MS or ESI MS in
3
neither positive nor negative modes. λmax (DMF, for tetra sodium salt 2, 1 µM)/nm 718 (ε/dm
ꢀ1
ꢀ1
mol cm 30 900), 680 (28 000), 348 (26 200).
Prepration of 1,8,15,22-tetrakis(N,N,N-diethylmethylammonioethylsulfanyl)-
,4,10,11,17,18,24,25-octamethyltetra(3,4-pyrido)porphyrazinato zinc(II) tetraiodide (4).
3
1,8,15,22ꢀtetrakis(N,Nꢀdiethylaminoethylsulfanyl)ꢀ3,4,10,11,17,18,24,25ꢀoctamethyltetra(3,4ꢀ
pyrido) porphyrazinato zinc(II) (3a) (80 mg, 66 ꢁmol) was dissolved in anhydrous DMF (2 mL)
under argon and methyl iodide (335 mg, 2.36 mmol) was added. The mixture was heated under
reflux for 24 h. The solvents were partially removed under a reduced pressure and acetone was
added. Compound was subsequently several times dissolved in MeOH and precipitated by
ACS Paragon Plus Environment
28

Page 29 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
diethylether. The green precipitate was collected and washed with diethylether, THF and acetone
to yield a dark green solid (78 mg, 67%). Anal. Calcd. for C H I N S Zn + 3H O: C, 41.76; H,
64 92 4 16
4
2
3
ꢀ1
ꢀ1
5
.37; N, 12.17; Found: C, 41.66; H, 5.25; N, 11.79; λmax (DMF, 1 µM)/nm 722 (ε/dm mol cm
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
11 400), 648 (46 900), 370 (44 500). IR (ATR) ν = 2978 (CH), 2934 (CH), 1555, 1476, 1404,
ꢀ1 1
1266, 1229, 1186, 1109 cm . H NMR (500 MHz; CD SOCD /pyridineꢀD ) δ = 4.74 – 3.69 (44
3
3
5
+
H, m, CH , ArCH ), 3.78 – 3.56 (12 H, m, N CH ), 3.20 – 2.93 (12 H, s, ArCH ), 1.53 (24 H, s,
2
3
3
3
1
CH CH ). H NMR (125 MHz; CD SOCD /pyridineꢀD ) δ = 157.77, 152.30, 151.01, 148.40,
2
3
3
3
5
142.24, 127.00, 125.09, 59.95, 56.56, 47.54, 23.44, 21.92, 17.13, 8.45, 8.45.
Preparation of 1,8,15,22-tetrakis(N,N,N-triethylammonioethoxy)-2,9,16,23-tetraethyl-
3
,4,10,11,17,18,24,25-octamethyltetra(3,4-pyrido)porphyrazinato zinc(II) octaiodide (5).
Magnesium (207 mg, 8.5 mmol) and a small crystal of iodine were refluxed in freshly distilled
anhydrous butanol (10 mL) until all Mg was converted to magnesium butoxide (approx. 4 h).
Compound 11 (368 mg, 1.35 mmol) was added and the reflux continued for another 15 h. After
cooling, the solvent was evaporated, and the product was extracted to THF and filtered. The
filtrate was concentrated to a small volume (approx. 5 mL), and precipitated by addition of
acetone:hexane (1:1, 100 mL). The precipitate of the Mg complex was collected as a dark green
solid, washed with acetone and characterized by MALDIꢀTOF MS (m/z 1040.6 [M –
+
N(C H ) ] ). The Mg complex was directly converted to the metalꢀfree ligand upon dissolution
2
5 2
in 1% HCl (50 mL). The mixture was stirred at rt for 2 h, at which time the reaction was
neutralized with 10% aq. NaOH and the product precipitated. The green solid of the metalꢀfree
+
+
ligand was collected and washed with water (m/z 1018.5 [M – N(C H ) ] , 1091.6 [M + H] ).
2
5 2
The whole product was subsequently dissolved in pyridine (7 mL) and refluxed for 4 h with zinc
acetate (110 mg, 0.6 mmol). The pyridine was evaporated, water was added with addition of
ACS Paragon Plus Environment
29