Photobleaching Efficiency Parallels the Enhancement of Membrane Damage for Porphyrazine Photosensitizers

Article abstract of DOI:10.1021/jacs.9b05991

Photostability is considered a key asset for photosensitizers (PS) used in medical applications as well as for those used in energy conversion devices. In light-mediated medical treatments, which are based on PS-induced harm to diseased tissues, the photoinduced cycle of singlet oxygen generation has always been considered to correlate with PS efficiency. However, recent evidence points to the fundamental role of contact-dependent reactions, which usually cause PS photobleaching. Therefore, it seems reasonable to challenge the paradigm of photostability versus PS efficiency in medical applications. We have prepared a series of Mg(II) porphyrazines (MgPzs) having similar singlet oxygen quantum yields and side groups with different electron-withdrawing strengths that fine-tune their redox properties. A detailed investigation of the photobleaching mechanism of these porphyrazines revealed that it is independent of singlet oxygen, occurring mainly via photoinduced electron abstraction of surrounding electron rich molecules (solvents or lipids), as revealed by the formation of an air-stable radical anion intermediate. When incorporated into phospholipid membranes, photobleaching of MgPzs correlates with the degree of lipid unsaturation, indicating that it is caused by an electron abstraction from the lipid double bond. Interestingly, upon comparing the efficiency of membrane photodamage between two of these MgPzs (with the highest and the lowest photobleaching efficiencies), we found that the higher the rate of PS photobleaching the faster the leakage induced in the membranes. Our results therefore indicate that photobleaching is a necessary step toward inflicting irreversible biological damage. We propose that the design of more efficient PS for medical applications should contemplate contact-dependent reactions as well as strategies for PS regeneration.

Full text of DOI:10.1021/jacs.9b05991

Subscriber access provided by Karolinska Institutet, University Library
Article
Photobleaching efficiency parallels the enhancement of
membrane damage for porphyrazine photosensitizers
Thiago T Tasso, Jan C. Schlothauer, Helena C Junqueira, Tiago A Matias, Koiti Araki, Erica
Liandra-Salvador, Felipe C.T. Antonio, Paula Homem-de Mello, and Maurício S. Baptista
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b05991 • Publication Date (Web): 06 Sep 2019
Downloaded from pubs.acs.org on September 7, 2019
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 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 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.
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 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Photobleaching efficiency parallels the enhancement of membrane
damage for porphyrazine photosensitizers
Thiago T. Tassoa*, Jan C. Schlothauera, Helena C. Junqueiraa, Tiago A. Matiasb, Koiti Arakib, Érica
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
Liandra-Salvadorc, Felipe C. T. Antonioc, Paula Homem-de-Melloc and Mauricio S. Baptistaa*
aDepartment of Biochemistry, Chemistry Institute, University of São Paulo, São Paulo, 05508-000, Brazil.
bDepartment of Fundamental Chemistry, Chemistry Institute, University of São Paulo, São Paulo, 05508-000, Brazil.
cCenter of Natural Sciences and Humanities, Federal University of ABC, Santo André, 09210-580, Brazil.
ABSTRACT: Photostability is considered a key asset for photosensitizers (PS) used in medical applications as well as for
those used in energy conversion devices. In light-mediated medical treatments, which are based on PS-induced harm to
diseased tissues, the photoinduced cycle of singlet oxygen generation has always been considered to correlate with PS
efficiency. However, recent evidence points to the fundamental role of contact-dependent reactions, which usually cause
PS photobleaching. Therefore, it seems reasonable to challenge the paradigm of photostability versus PS efficiency in
medical applications. We have prepared a series of Mg(II) porphyrazines (MgPzs) having similar singlet oxygen quantum
yields and side groups with different electron-withdrawing strengths that fine tune their redox properties. A detailed
investigation of the photobleaching mechanism of these porphyrazines revealed that it is independent of singlet oxygen,
occurring mainly via photoinduced electron abstraction of surrounding electron rich molecules (solvents or lipids), as
revealed by the formation of an air-stable radical anion intermediate. When incorporated into phospholipid membranes,
photobleaching of MgPzs correlates with the degree of lipid unsaturation, indicating that it is caused by an electron
abstraction from the lipid double bond. Interestingly, upon comparing the efficiency of membrane photodamage between
two of these MgPzs (with the highest and the lowest photobleaching efficiencies), we found that the higher the rate of PS
photobleaching the faster the leakage induced in the membranes. Our results therefore indicate that photobleaching is a
necessary step towards inflicting irreversible biological damage. We propose that the design of more efficient PS for medical
applications should contemplate contact-dependent reactions as well as strategies for PS regeneration.
mechanistic alternatives.7 Synergistically coupling 1O2
production with other actions such as the release of
chelating agents, coupling of PS to proteins favoring type I
reactions, or developing photochemotherapeutic agents
that work under low oxygen concentrations, are examples
of novel promising strategies in the area of
photomedicine.8–10
INTRODUCTION
Photobleaching causes
a
loss of absorption of
chromophores under light exposure. Therefore, many
phenomena that require light absorption and the
subsequent photoinduced action of photosensitizers (PSs),
such as natural or artificial photosynthesis, dye-sensitized
photovoltaic cells and optogenetics, suffer from
photobleaching.1–3 Accordingly, PSs are usually designed to
experience the minimum possible level of photobleaching.
The same principles have been applied in photomedicine
in the development of PSs against proliferative or infective
diseases. However, there is no definitive evidence showing
that photobleaching is as detrimental in photomedicine as
it is in other areas.4 Indeed, it is commonly accepted that
the more stable the PS the better it will perform, mainly
because it can endure more cycles of singlet oxygen (1O2)
production and efficiently trigger the consequent type II
oxidation processes.5
For most PSs, photobleaching proceeds via an oxygen-
1
dependent pathway, i.e., involving the intermediacy of O2,
which can add to macrocycle double bonds.11 Arnaut and
coworkers have shown that photostability can be enhanced
by attachment of electron-withdrawing groups to the
macrocycle, increasing the oxidation potential and
imposing a barrier for 1O2 attack.12,13 Other PSs photobleach
through direct-contact reactions of their excited states
with surrounding molecules. In principle, these reactions
can cause the neutralization of a specific biological target,
but there is still a lack of knowledge about their
mechanisms and consequences.14,15 For example, Bacellar
and co-workers have showed that PSs have to be physically
bound to the membranes in order to cause membrane
leakage.16 The generation of membrane pores was shown to
depend on the accumulation of lipid truncated aldehydes,
Although hundreds of “improved” photosensitizers have
been planned and prepared based on these premises, it has
been challenging to develop considerably more efficient
Photodynamic Therapy (PDT) protocols based solely on
this principle.6 Therefore, it is imperative to look for other
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
which are formed by the PS-induced electron abstraction
Page 2 of 13
RESULTS AND DISCUSSION
1
2
3
4
5
6
7
8
either directly from the double bond of the lipid or from
hydroperoxides that were previously generated by the ene
reaction (Scheme 1). Since an electron transfer reaction
between the PS and the biological target can cause PS
photobleaching, we put forward the hypothesis that
photobleaching can be beneficial to the outcome of the
photosensitized oxidation reactions.
Synthesis
and
photophysical
studies.
The
magnesium(II) porphyrazines (Pz) shown in Figure 1 were
synthesized in a single-step by cyclization of the respective
dinitriles (cis/trans mixture) in the presence of Mg(OBu)2
under reflux. While the meta-substituted phenyl rings of
CF3Pz confer high solubility in most organic solvents, the
para-substituted Pzs are more symmetric and thus soluble
mainly in THF. In this sense, CF3Pz was purified by column
chromatography, while the purification of the other Pzs
was achieved by successive recrystallizations from
9
Scheme 1. Oxidation pathways of unsaturated lipids
via contact-dependent and independent reactions
between the PS and the lipid.
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
1
THF/MeOH. H NMR, 13C NMR and MALDI-TOF mass
spectrometric analysis were used for the characterization
of the starting materials and products, as shown in the
Supporting Information (SI). Substituents ranging from
strong electron-withdrawing (CF3) to electron-donating
groups (MeO) were chosen in order to modulate the Pz
electronic properties.
The Pzs showed characteristic * absorption bands in
the red (Q band,  ~ 105 M-1cm-1) and blue regions (Soret
band) of the spectrum, with MeOPz showing an additional
band at 490 nm, arising from a charge transfer transition
from the substituents to the porphyrazine ring (Figure S1
in the SI).17 The Q band position red-shifts from 630 nm for
CF3Pz to 649 nm for MeOPz following the order CF3 < F <
Cl < Br < MeO, with the same trend observed for the
fluorescence emission maxima (Figure 2, Table 1). The
values of the Stokes shifts (ca. 17 nm) are in agreement with
those found for other porphyrazines18,19, and reflect the
high rigidity of the macrocyclic ring. Pzs are usually
efficient fluorophores, but susceptible to the influence of
the substituents. Fluorescence quantum yields (F) varied
from 0.18 for CF3Pz to 0.05 for BrPz, because of the
enhancement of the intersystem crossing imposed by the
bromine atoms, as evidenced by the increase in the singlet
oxygen production (see below). F was even smaller for
MeOPz (0.007) due to an alternative route of excited state
deactivation, which is a photoinduced charge transfer from
the peripheral groups to the Pz ring (see below).
In this work, we aim to test this hypothesis with well-
controlled model systems. We designed a series of
magnesium porphyrazines (Figure 1) substituted with
symmetrically distributed electron-donating or electron-
withdrawing phenyl groups, in order to obtain PSs with
tunable levels of photostability. We initially showed that
the photobleaching mechanism of these molecules is
1
independent of O2 and involves an electron abstraction
from surrounding molecules. Importantly, by comparing
the photoinduced disruption of phospholipid membranes
with the photobleaching rate of the PSs, we challenge the
paradigm that photobleaching is necessarily correlated
with low PS efficiency.
To define the HOMO and LUMO gap for this Pz series,
TDDFT calculations using the B3LYP functional were
performed. The HOMO and LUMO are centered on the
porphyrazine ring and their energy gap correlates well with
the red shift order observed in their solution spectra
(Figure S2). Electron-donating groups tend to destabilize
both the HOMO and LUMO of the macrocycle. However,
a stronger destabilization of the HOMO is responsible for
the decrease in the energy gap from CF3Pz to MeOPz,
explaining the consequent red shift in the spectrum.
Figure 1. Molecular structure of the MgPzs.
2
ACS Paragon Plus Environment

Page 3 of 13
Journal of the American Chemical Society
1
2
3
Table 1. Photophysical parameters obtained for the
Pzs in THF solution.
4
5
6
7
8
9
10
11
12
emiss
nm
/

abs / nm
a
b
Pzs
F

(log 
CF3Pz
FPz
630 (4.96)
634 (4.87)
638 (4.96)
639 (4.92)
648
651
0.177 ± 0.001
0.110 ± 0.001
0.11 ± 0.01
0.34 ± 0.07
0.34 ± 0.02
0.48 ± 0.04
0.64 ± 0.06
0.07 ± 0.01
ClPz
BrPz
655
656
666
0.053 ± 0.005
0.007 ± 0.002
MeOPz 649 (5.03)
a Cresyl violet in MeOH as reference (F = 0.57).b Methylene
blue ( = 0.52) and 1,9-dimethylmethylene blue ( = 0.71) in
EtOH as references.
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 2. Normalized absorption (solid lines) and
fluorescence emission (dashed lines) spectra of CF3Pz (black),
FPz (red), ClPz (blue), BrPz (pink) and MeOPz (green) in
THF.
Photostability studies were conducted by irradiating and
following the absorbance change over time of the Pzs
(Figures 3a, S4). Photobleaching was irreversible and was
characterized by absorbance loss in all Pz bands (300 to
700 nm), accompanied by a small absorbance increase in
the 700 to 800 nm region (Figure 3a). Absorption of Pzs in
the NIR have been previously attributed to a macrocycle-
based radical species (this will be further discussed
below).24 Photobleaching quantum yields (pb) of Pzs were
calculated by using the values of the photobleaching rate
(kPB) taken from the fitting of the absorption decay curves
to a kinetic model, which was developed to allow better
fittings to the experimental data (see SI – section 1.1). As
clearly seen in Figure S5, the data fit with the kinetic model
is a lot better than the fit using a mono-exponential
function. It is worth noting that photobleaching varied
substantially within the Pz series (Figure 3b). pb values
decreased with the increase in the electron donating ability
of the substituents, as seen in Table 2. Using the Hammett
sigma parameter (), which accounts for both inductive
and resonance effects of substituent groups,25 it was
Pz macrocycles usually show high singlet oxygen quantum
yields (), which justifies their use in medical
applications.  of Pzs were obtained by measuring the 1O2
phosphorescence decay at 1270 nm (Figure S3) and
comparing to MB and DMMB in ethanol as references (see
SI). The values (Table 1) clearly followed the trend based
on the presence of heavy atoms, increasing from FPz (0.34)
to BrPz (0.64), which showed the highest yield among the
Pzs. The efficiency was lowest for MeOPz (0.071), in
agreement with its excited states being deactivated mainly
by non-radiative pathways.
Photobleaching. There are many photobleaching
mechanisms that can be generically classified as oxygen-
dependent and -independent mechanisms. Oxygen
dependent photodegradation usually involves oxidation by
1O2 and is common for porphyrinoids12,13,20 and other
chromophores such as BODIPY21. The attachment of
electron-withdrawing groups to the ring increases its
oxidation
potential,
resulting
in
a
greater
photostability12,13,21. The oxygen-independent mechanism
usually involves direct-contact reactions between the
excited state of the PS and neighboring molecules, which
can be the dye itself (dye-dye mechanisms)14,22 or
preferably a biological target that can be key to the cellular
response.23 An initial electron or hydrogen transfer
reaction leads to the formation of radicals that undergo
further reactions. Although very likely to occur, less
attention has been paid to these reactions compared with
the oxygen-dependent photodegradation mechanisms.
possible to find
a linear relationship between the
substituents electronic nature and the porphyrazine
photostability, as shown in Figure 3c. A higher  value
indicates a more electron-withdrawing (or less electron-
donating) nature of the group, so, clearly, CF3 is the most
electron-withdrawing group of the series. Therefore, if one
wants to minimize photobleaching of the Pz, one should
choose electron-donating substituents, suggesting that an
electron donation to the macrocycle is involved in the
photo-induced degradation of the Pzs. MeOPz is an
outlier since it showed no detectable photobleaching
under the conditions used in the experiment. Although it
has the strongest electron donating group in the series, the
absence of photodegradation cannot be solely explained by
the redox properties of the excited state, but it is mainly a
consequence of the intramolecular suppression
mechanism of the excited state, which is in accordance
with the photophysical data shown above (see Figure S6 for
additional discussion).
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 13
(Table S1). A possible involvement of residual water in the
1
2
3
4
degradation mechanism was discarded, because the
photobleaching rates in anhydrous (water content <
0.002%) and water-containing solvent were essentially the
same (Figure S9).
5
6
7
8
pb was plotted as
a
function of polarity and
nucleophilicity, showing no clear correlation (Figure S10),
suggesting that a nucleophilic attack by the solvent on the
Pz ring is not a probable mechanism involved in
photodegradation. However, good correlations are
9
observed with Gutmann’s donor number26 and the
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
coordination ability (aTM
)
of the solvents (Figure 4b),
27
parameters that reflect the interaction between a Lewis
base (donor) and a Lewis acid (acceptor). In the work of
Mielcarek and collaborators,28 the photobleaching
reported for free-base (2H) amino-substituted Pzs was 70
times slower than that for the Zn(II) and Mg(II) complexes.
The authors gave no explanation for these observations,
nor for the strong photobleaching observed. However,
their results reinforce that solvent coordination to the
central metal can help promote the electron transfer
process, increasing photobleaching. It is indeed likely that
solvents with high donor number strongly coordinate to
the Mg(II) center of the porphyrazine, stabilizing the
donor-acceptor pair for the electron transfer reaction to
proceed. Toluene, for instance, shows a very weak
coordination ability and thus the driving force must be
highly endergonic, explaining the very low rate of
photobleaching in this solvent. Interestingly, the
photostability of Pzs in biological environments was also
studied by inserting them into phospholipid vesicles.
When incorporated in vesicles of DMPC, a phospholipid
containing only saturated chains, the pb values decreased
sharply for all Pzs, as shown in Table 2 (data with other
lipids are shown below).
Figure 3. (a) Spectral changes for CF3Pz in air-saturated THF
solution under Q band irradiation. (b)Absorbance decay with
time upon irradiation at the Q band for MeOPz (green), FPz
(red), ClPz (blue), BrPz (pink) and CF3Pz (black) in THF. (b)
Correlation between the photostability of the Pzs and the
Hammett sigma value of the substituents (m for the meta-
substituted CF3Pz and p for the para-substituted Pzs).
Pzs photodegradation is not affected by oxygen, i.e., when
oxygen is removed, pb values are very similar to those
observed in air-saturated solutions (Table 2), even though
oxygen quenches Pzs triplets efficiently (Figure S7).
Therefore, the results in the presence of oxygen rule out
the involvement of triplet species and suggest that the
photodegradation occurs directly in the singlet excited
state by an electron-transfer reaction with molecules that
are in the solvation sphere of the Pzs singlet excited state.
Note also that the triplet-triplet absorption of CF3Pz does
not return to zero, in either air saturated or argon purged
Classical photolysis data suggest that Pzs photobleaching
takes place through a reductive mechanism by accepting
an electron from neighboring donors (D), without
interference of type II mechanism or dye-dye reactions.
Thus, the fundamental step is the photoinduced formation
of the Pz radical anion, which was previously detected by
spectroscopic techniques.23 Interestingly, electron
paramagnetic resonance (EPR) measurements of irradiated
pyridine and THF CF3Pz solutions showed the presence of
a well-defined singlet signal with g = 2.03, absent in the
non-irradiated solutions (Figure 5, S11). This signal, with no
hyperfine coupling, is characteristic of a single electron
radical, which is delocalized over the porphyrazine ring.29
Table S2 contains the values of spin density calculated for
all MgPzs, showing that the free electron is delocalized
over the whole Pz ring, but with no charge density in the
magnesium ion.
solutions, indicating the formation of
a long-lived
intermediate species upon excitation (Figure S7). Besides,
the fact that the photodegradation rates follow the first-
order kinetic model, as calculated for CF3Pz in THF (Figure
S8), indicates that Pzs do not tend to engage in dye-dye
photobleaching mechanisms (dye-dye mechanisms would
require second or higher order kinetics).
In order to further understand the mechanism of
photodegradation of Pzs, photobleaching rates of CF3Pz
were obtained in different organic solvents, taking
advantage of its improved solubility compared to the other
Pzs. As shown in Figure 4a, the photobleaching rate of
CF3Pz increases in the order toluene < acetone <
acetonitrile < THF < pyridine. From toluene (pb = 4.2 x 10-
6) to pyridine (pb = 4.4 x 10-4), the pb values increased by
two orders of magnitude, indicating a strong solvent effect
4
ACS Paragon Plus Environment

Page 5 of 13
1
2
3
4
5
6
7
Journal of the American Chemical Society
experiment led to an apparent degradation of the
macrocycle into smaller, less conjugated fragments as
indicated by depletion of the NIR peaks and appearance of
new bands at 412 and 438 nm (Figure S13), suggesting that
the absence of the oxidation peak is caused by subsequent
chemical reactions.30 These results are in agreement with
those reported by Schröder et al. for a resonance-stabilized
organic radical anion.30 Although this anion radical is
stable both in the solid state and in solution, the precursor
molecule also shows a single irreversible electrochemical
reduction in cyclic voltammetry measurements.30 MALDI-
TOF-MS spectra for both irradiated and non-irradiated
solutions of CF3Pz are rather similar, and the high
abundance of the molecular ion species [M] (m/z 1488.1)
in the negative mode (Figure S14) corroborates with the
proposed CF3Pz structure as the main photobleaching
product for this Pz. Theoretical calculations were used to
obtain additional information about this intermediate
species. One-electron reduction of the Pz ring was shown
to break the degeneration of the HOMO-LUMO transition,
i.e., Q band splits into two transitions (one higher and
other lower in energy – Table S3 and Figure S15), agreeing
with the absorbance changes around 700 nm observed
during photobleaching and electrochemical reduction of
CF3Pz (Figure 5b).
Table 2. Photobleaching quantum yields (pb x 105) for
the Pzs in THF solution and in DMPC liposomes.
THF
(air)
THF
(argon)
DMPC
(air)
Pzs
CF3Pz
FPz
15 ± 2
14 ± 3
2.6 ± 0.1
6.1 ± 0.9
6 ± 3
1.3 ± 0.3
1 ± 1
2.5 ± 0.1
3.9 ± 0.2
4.8 ± 1
**
8
9
ClPz
0.8 ± 0.4
0.7 ± 0.1
**
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
BrPz
MeOPz
**
Irradiation at the Q band (74 mW, pulse frequency of 10
Hz).**No photobleaching was detected under the
experimental conditions used in this work.
The stability of CF3Pz●− was tested in aerated THF solution
by monitoring its absorption band (ca. 700 nm) over time.
This species is stable for more than one week in solution,
even in the presence of water in the medium (Figure S16).
Molecular oxygen (Ered = 0.33 V)31 does not seem to oxidize

Figure 4. (a) Absorbance decay with time upon irradiation in
the Q band of CF3Pz in pyridine (black), THF (pink),
acetonitrile (blue), acetone (red) and toluene (green). (b)
Relationship between the CF3Pz photobleaching quantum
yield (pb) and the solvent donor number and coordination
ability (inset).
the Pz radical to form superoxide anion radical (O2 ) since
CF3Pz●− is stable in the presence of oxygen and
regeneration of the original macrocycle was never
observed.
2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ, Ered = +0.50 V)32, a stronger oxidant compared with
molecular oxygen, could not oxide the CF3Pz radical either.
Ceric ammonium nitrate (CAN, Ered = +1.70 V)33, however,
completely depleted the band around 700 nm (Figure S17),
indicating that with enough oxidizing potential, it is
possible to oxidize CF3Pz. Since no bands appeared in the
visible region, the oxidation of CF3Pz species must lead
to degradation of the Pz ring (resulting in smaller products
with little or no conjugation). These data are also in
agreement with the spectroelectrochemical results above.
Surprisingly, this EPR peak is stable at room temperature
even in the presence of oxygen (Figure S11), explaining the
relatively long-lived species observed in the transient
absorption measurements discussed above (Figure S7).
Further
electrochemical
investigations
(spectroelectrochemical and cyclic voltammetric) agrees
with this interpretation. Indeed, when a reduction
potential of -0.7 V (vs. Ag/AgNO3 10 mM in acetonitrile)
was applied to a CF3Pz solution in pyridine containing 0.1
In summary, several data from different techniques
(transient absorption, EPR, spectroelectrochemistry
MALDI-TOF-MS and theoretical calculations) all agree
with the formation of a π-radical Pz derivative species
(MgPz) as the intermediate in the photodegradation
pathway. Although porphyrin and phthalocyanine radicals
tend to be extremely unstable in solution (especially in
aerated environments)34, MgPz shows up to be fairly
stable even in the presence of oxygen. This long-lived
radical species is interesting especially to other research
areas, such energy conversion devices, and certainly
deserves further investigation. However, we consider that
M
tetrabutylammonium perchlorate (TBAP), its
absorption spectrum started to change in a way similar to
that observed during irradiation of the solution at 630 nm,
as shown in Figure 5. All typical absorption bands of CF3Pz
decreased in intensity, followed by the surge of new
absorption bands at 702 and 765 nm. This process is
irreversible, in agreement with the cyclic voltammetry
measurement of CF3Pz in deaerated pyridine, which shows
a single irreversible reduction process at -0.9 V vs.
Ag/AgNO3 (Figure S12). Oxidation of the reduced species
by sweeping back to -0.2V in the spectroelectrochemistry
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 13
further studies with MgPz are out of the scope of this
paper.
experiments of liposomes having membrane-incorporated
1
2
3
4
5
6
7
8
CF3Pz shows that triplets are not suppressed by the lipid
double bonds (Figure S20), in agreement with the electron
abstraction step being a consequence of the direct reaction
with the singlet excited state of CF3Pz, as observed in bulk
solutions (Table 2). These results clearly show the
influence of biological targets (lipid double bonds) on the
photodegradation rate of these compounds. The feasibility
of a photoinduced electron-transfer between the lipid
unsaturation and the CF3Pz excited state is discussed in
terms of driving force (G) in the SI section.
The trend observed in the plot of Figure 3c can now be
rationalized based on this proposed mechanism. Electron-
donating groups attached to the Pz ring considerably
increase the energy of the Pz’s LUMO (see Figure S2) and
decrease electron affinity (Table S2), which disfavors the
reduction of the macrocyle thermodynamically. The
opposite is observed for substituents that withdraw
electronic density from the ring, explaining the higher
photobleaching rate observed for CF3Pz. Indeed, the spin-
orbital energy calculated for CF3Pz is the lowest (~−4eV)
in the MgPzs series (Figure S18), indicating that the
trifluoromethyl groups largely stabilize the additional
electron.
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 5. Changes in the absorption and EPR spectra (inset)
of CF3Pz in pyridine before (black line) and after (red line)
application of -0.7V to -1.0V (a) or irradiation at 632 nm for 20
min (b).
Photostability vs. membrane damage efficiency. The
photoinduced efficiency of cell death induced by a PS is
often associated with membrane damage, since subtle
damages in membranes, either the cytoplasmic membrane
or those of intracellular organelles, have important
consequences for cells, defining survival or death
outcomes.7,35 In order to correlate membrane damage with
photobleaching efficiency, we performed experiments in
several membrane mimic systems. First, we investigated
the photobleaching rates of CF3Pz in the presence of lipids
with different degrees of unsaturation and compared them
with the rate of membrane leakage. As shown in Figure 6
and Table S4, the rate for CF3Pz photobleaching in POPC
(a monounsaturated lipid) is about two orders of
magnitude higher than in DMPC (a saturated lipid) under
the same conditions, indicating that the unsaturated C=C
bond act as electron donor in the photoinduced redox
process that lead to Pz bleaching. This is further confirmed
by the rate in DOPC (lipid with two unsaturation sites, one
in each carbon chain), which showed a ca. three times
higher value than in POPC vesicles. Note also that
irradiation of CF3Pz in the presence of unsaturated lipids
leads to the formation of CF3Pz in both air saturated and
deaerated solutions (Figure S19). Triplet-triplet absorption
Figure 6. Photobleaching kinetics for CF3Pz (0.5 mol%)
incorporated in the membrane of liposomes of (a) DMPC,
POPC and (b) DOPC under irradiation with red LEDs (620 –
625 nm).
Aiming to correlate photobleaching with membrane
leakage, we compared the membrane damage efficiencies
of two PSs (CF3Pz and FPz) with similar photophysical
properties (absorption maxima, molar absorption
1
coefficients, fluorescence and O2 quantum yields − Table
1), but distinct photobleaching efficiencies (Table 2). The
results of the membrane damage experiments are shown in
Figure 7. When liposomes containing 0.25 mol% of Pz were
irradiated, liposome-entrapped carboxy-fluorescein (CF)
release was clearly faster for CF3Pz than for FPz. The
release rate, taken from the linear fit of the curves from 0
to 30 min, was ca. 4 times higher for CF3Pz (3.1%/min)
compared to FPz (0.8%/min). The same experiment was
6
ACS Paragon Plus Environment

Page 7 of 13
Journal of the American Chemical Society
reproduced with a very low Pz concentration in the membrane leakage and in agreement with previous
literature.16 By using a series of Pz molecules, our data
clearly indicate that PSs having faster bleaching are more
effective at causing irreversible damage in membranes
(Figure 8). CF3Pz is able to oxidize molecules, not only by
the action of diffusing species such as 1O2, O2 etc, but also
via direct reaction with suitable target molecules located
within the contact volume. While the reaction with oxygen
is basically dependent on the velocity of diffusion of the
reactants (quantum yields in the order of 0.3-0.6), the rate
of type I contact-dependent oxidation of the double bond
depends on the presence of the photosensitizer in physical
contact with the target and in the proper geometry. This
results in comparably lower photobleaching quantum
yields (10-4-10-5). Even though being relatively rare, these
contact-dependent reactions are the ones that trigger
membrane leakage. In the case of the Pz, it is likely that the
contact-dependent reaction occurs within the solvation
sphere of the PS (no diffusion), since it occurs within the
1
2
3
4
5
6
7
8
membrane (0.06 mol%) in order to follow the absorption
changes of the PSs (Figure 7b and inset). During the first
100 min of experiment, the rate of CF release was similar
for both Pzs since the Q band absorbance values indicated
no significant bleaching until this point. Proceeding with
the irradiation, CF3Pz outperforms FPz in terms of causing
membrane leakage and it is possible to see a marked
decrease of the CF3Pz Q band absorbance with an increase
of the 700 nm absorption, indicating formation of the Pz
species. Therefore, the higher photobleaching rate of
CF3Pz improves its efficiency in causing membrane
leakage.
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
In order to visualize the progress of membrane oxidation
and leakage, CF3Pz was incorporated into giant
unilamellar vesicles (GUVs) of DOPC loaded with sucrose
in a glucose medium and monitored by phase-contrast
microscopy. As shown in Figure 7c, in the first minutes of
irradiation, the shape of the GUV is highly deformed as the
membrane tries to accommodate the oxidized lipid formed
via the photosensitization by CF3Pz. Membrane buds are
then formed and expelled from the GUV, while it resumes
its circular shape with smaller diameter, due to lipid loss.
From this point forward, the membrane starts to lose its
contrast, indicating pore formation and exchange between
the inner content (sucrose) and the external solution
(glucose). The progress of phase contrast loss was
monitored as a function of time (Figure S21) and, after 34
min of irradiation, the GUV membrane was only barely
visible. FPz also induces membrane fluctuations and bud
formation in the first few minutes of irradiation, but
contrast loss was not observed even after 35 min, indicating
no membrane leakage (Figure 7d). This agrees with the
hypothesis that PS needs to photobleach in order to make
a membrane leak and those that photobleach faster are the
ones that cause faster irreversible membrane damage.
lifetime of the singlet excited state
(~1 ns)38. It is
interesting to observe that membrane binding is not the
only factor affecting membrane leakage, since other PSs
that are fully integrated into membranes, but only generate
1O2, do not cause membrane permeation.39 Porphyrazines
do so because they are stronger oxidizing agents and
consequently able to abstract an electron from the lipid
double bond.40,41 Nevertheless, this concept can now be
used to design and develop optimized PSs, which ideally
could specifically oxidize and inactivate intracellular
molecular targets (Figure 8).
CONCLUSIONS
A series of Mg(II) porphyrazine complexes with similar
photophysical and tunable redox properties was prepared
and studied both in bulk solution and in membrane
mimetic systems. PS photobleaching was shown to occur
by the oxidation of neighboring molecules, leading to the
formation of a PS-derived anion radical species. Within
membranes, PS photobleaching occurs through an
electron abstraction from the lipid double bond, which
causes irreversible membrane damage. Therefore, our data
support a re-conception of the role of photobleaching in
PDT by showing a positive correlation between the
efficiencies of PS photobleaching and membrane leakage.
Given an intracellular target, contact-dependent reactions
can damage the biological target with a lot more precision
compared with diffusive species such as 1O2 or other ROS.
In addition, the use of drugs that have defined
photobleaching rates can be crucial for the development of
dosimetry in hypoxic regions, as well as for the clearance
of the drug from the body.42,43 We hope our work will
stimulate the development of new lead compounds for
PDT, by designing molecules (e.g. metal complexes,
phenothiazines, among others) with improved redox
properties and also considering strategies that allow PS
replenishment.
PS photobleaching was never thought to improve
photoactivity,28-31 instead, improved PS efficiency has been
usually associated with high 1O2 quantum yields. It is well
documented though that 1O2 reacts with unsaturated lipids
to form hydroperoxides, which alters membrane area and
thickness but does not induce its permeabilization.36,37
Bacellar et. al. showed that membrane permeabilization is
dependent on the accumulation of truncated lipid
aldehydes, which are only formed by the direct-contact
reactions of the excited state photosensitizer with either
the lipid double bond or a lipid hydroperoxide.16 By
abstracting electron/hydrogen from these targets, the PS is
able to form peroxyls and alkoxyls radicals leading the
carbon-carbon bond breakage and aldehyde formation.16
We also searched for lipid oxidation products originated
from contact-dependent reactions with CF3Pz. As shown
in Figure S22, liposomes incorporated with this PS induce
much higher accumulation of malondialdehyde (MA),
which is a short-chain aldehyde formed by C-C chain
breakage of the phospholipid, compared to a control
liposome (without Pz), explaining the higher rate of
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 13
deactivated to the ground state; be converted to the triplet
1
2
3
4
5
6
7
8
state (T1) via intersystem crossing (ISC); or engage in an
electron-transfer process with surrounding molecules as
solvent (solv) and unsaturated lipids. This irreversible process
leads to photobleaching due to formation of a long lived
reduced Pz species (Pz), which has low visible light
absorption. The oxidation of lipids by Pzs through this
contact-dependent reaction leads to the formation of
membrane pores, resulting in membrane leakage. Therefore,
photobleaching is
a necessary step towards inflicting
9
irreversible biological damage.
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
EXPERIMENTAL SECTION
Materials and methods. Diethyl ether and
tetrahydrofuran were refluxed over sodium and distilled
prior to use for synthesis and photophysical
measurements. All the other solvents and reagents were
used without further purification. The phenylacetonitrile
precursors were purchased from Sigma-Aldrich; 1-
palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC),
1,2-Dimyristoyl-sn-glycero-3-phosphocholine (DMPC) and
1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) were
purchased from Avanti Polar Lipids; and the solvents were
analytical grade and acquired from Synth.
1H NMR measurements were performed in a Bruker Avance
III equipment operating at 300 MHz. For characterization
of the porphyrazines, mass spectra were recorded in a
Autoflex Speed-MALDI-TOF/TOF equipment in the
positive mode, using 2,5-dihydroxybenzoic acid as matrix.
Absorption spectra in the UV-Vis region were recorded in
a Shimadzu UV-1800 spectrophotometer. Fluorescence
spectra were recorded from 580 to 800 nm in a Cary
Eclypse fluorometer, and the quantum yields were
Figure 7. Plots of 5(6)-carboxyfluorescein release (%) from
DOPC liposomes containing no Pz (black squares), CF3Pz (red
circles) or FPz (blue triangles) at (a) 0.25 mol% and (b) 0.06
mol% under red light irradiation (max = 630 nm). Inset:
Variation of the absorbance of the CF3Pz Q band (red circle)
and 700 nm band (red star) and of the FPz Q band (blue
triangle) and 700 nm band (blue cross) in the liposomal
suspension during the experiment. Images of a DOPC GUV
with CF3Pz (c) and FPz (d) 1 mol% with different times of
irradiation with blue light (440-495 nm).
obtained using cresyl violet as reference (F = 0.57 in
MeOH)44 according to the procedure described in ref.
.
45
The samples were excited at 570 nm, with excitation and
emission slits of 5 nm. Singlet oxygen measurements were
performed
in
an
Edinburgh
time-resolved
Hamamatsu
phosphorescence spectrophotometer.
A
R5509-72 photomultiplier detects the 1O2 phosphorescence
at 1270 nm. The samples were excited at 630 nm using a
OPO RAINBOW laser system from Quantel coupled to a
Magic PrismTM, which can convert the Nd:YAG laser
wavelength into a continuously tunable emission from 420
1
to 680 nm. The O2 quantum yields () were calculated
from the equation below using the amplitudes (APz) of the
mono-exponential decay function fitted to the 1O2
phosphorescence decay curves of each sample in THF,
correlating with the quantum yields and respective values
(Aref) for methylene blue (
=
0.52)46 and 1,9-
dimethylmethylene blue (DMMB,  = 0.71)47 in EtOH as
references. By using the amplitude instead of the area
under the decay curves, it is possible to minimize the error
associated with using different solvents for the samples
and references.
Figure 8. Major photo-induced processes of the MgPzs. The
MgPz excited state formed upon light excitation (S1) can be
8
ACS Paragon Plus Environment

Page 9 of 13
Journal of the American Chemical Society
150 L of Tris buffer and 50 L of liposome suspension with
1
2
3
4
or without Pz (control). A final row containing 140 L of
Tris buffer, 50 L of liposome suspension and 10 L Triton
X-100 (10%) was prepared for calculation of the percentage
of release assuming these wells to represent 100% CF
release. Irradiation of the plate was done using a red LED
array (max = 630 nm) with irradiance of 6.6 mW/cm2. The
CF fluorescence emission was read in a microplate reader
(SpectraMax i3 from Molecular Devices) by exciting at 480
nm and reading the emission at 517 nm from bottom.
Detection of malondialdehyde in photosensitized lipid
membranes were performed according to the literature.49
Soy lecithin liposomes were prepared and irradiated
similarly to the membrane leakage assays for 2 hours. The
irradiated vesicles contained 0.5 mol% of CF3Pz or no
CF3Pz as a control.
퐴
∆푟푒푓
푃푧 
=
∆

퐴푟푒푓
5
6
7
8
For the photobleaching studies, a Pz solution with
maximum absorbance of 1.0 was irradiated at the Q band
(74 mW, 10 Hz pulse frequency, 5-10 ns pulse width) using
the OPO RAINBOW optical system described above, in 1
minute intervals for 10 to 20 minutes. At each interval, the
solution was homogenized and the absorbance spectrum
from 300 to 800 nm was registered. For better detection of
the photobleaching process, a quartz cuvette of 700 L
internal volume (1.0 x 0.4 cm) was used, with both
irradiation and absorbance measurements performed with
the 1.0 cm optical path. The photobleaching quantum
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
yields  푃퐵 were calculated using the equation below,
where:

kPB
Giant Unilamellar Vesicles (GUVs) were grown by the
electroformation method following a previously reported
procedure, using instead 0.2 M glucose and sucrose
solutions and 2 mM DOPC in CHCl3 containing 0.5% of
Pz.50 The GUVs were observed in a Zeiss Observer.D1
inverted microscope with phase contrast and 10x/40x
objectives (A-PLAN 10x/0.25 Ph1 and EC Plan-NEOFLUAR
40x/0.75). Images and videos were recorded with an
Axiocam MRm camera coupled to the microscope. A HXP
fluorescent lamp and Zeiss filter set 05 (excitation filter: BP
365-440nm; beamsplitter: FT 460 nm; emission filter: LP
470 nm) were used for irradiating the GUVs.
is the photobleaching rate in mol dm-3 s-1
calculated from the fitting of the photobleaching curves
(see SI for further details), V is the irradiated solution
volume in dm3, Np is the photon flux in einstein per
second, and Abs is the initial absorbance of the
porphyrazine Q band.
푘푃퐵
푁푝 (1 ― 10퐴푏푠
푉
=
푃퐵

)
The photon flux (Np) was calculated from actinometry
experiments using potassium tris(oxalato)ferrate(III) as
actinometer according to the procedure described in ref.
48. For the oxygen-free experiments, the cuvette containing
the Pz solution was bubbled with argon for 15 min and
Transient absorption measurements were performed in a
flash photolysis instrument (Applied Photophysics Ltd,
Surrey, United Kingdom) pumped by a nanosecond
Nd:YAG laser (Spectra Physics, Stahnsdorf,Germany) with
excitation wavelength at 355 nm. First-order decay kinetics
were observed for the lowest triplet state. Transient
absorption spectra (300–750 nm) were recorded by
monitoring the optical density change at 10 nm intervals
with signal averaging of 4 decays at each wavelength.
sealed with
a
septum prior to measurement.
Photobleaching of the Pzs in the lipid vesicles was
performed similarly to the procedure above, using instead
two red LEDs (620-630 nm), each one positioned in the
center of a transparent face of the quartz cuvette, with the
power supply set to 0.7A.
For
electron
paramagnetic
resonance
(EPR)
measurements, a solution of CF3Pz (0.5 M) in pyridine or
Lipid vesicles were prepared by evaporating a CHCl3
solution of 2.8 mg of DOPC, POPC or DMPC with the
MgPz (0.5% mol) under argon flow in a test tube. The films
were hydrated for 15 min in deionized water and the tube
was vortexed three times for 30 seconds or until complete
detachment of the film from the glass. The lipids were
extruded 11 times through a 100 nm pore membrane, and
used immediately for the photobleaching assays using the
same set up as for the solution experiments. For the
membrane leakage assays, 5(6)-carboxyfluorescein (50 mM
in 10 mM Tris buffer solution pH = 8.0) was encapsulated
by the same procedure described above, using instead it in
the solution for hydration of the lipid film. After extrusion,
non-encapsulated CF was removed by column
chromatography using Sephadex G-50 and Tris buffer (10
mM, NaCl 300 mM) as eluent. The liposome suspension
collected (ca. 2 mL) was used on the same day. In a Costar
96-well black polystyrene plate, each well was filled with
THF was transferred to a flat cell and the spectra were
recorded at room temperature on
a Bruker EMX
spectrometer equipped with a high sensitivity cavity and
operating at 9.85 GHz and 100 KHz field modulation. This
solution was then transferred to a cuvette (1.0 x 0.4 cm) and
irradiated at 630 nm for 20 min in the set up used for the
photobleaching studies. Immediately after irradiation, the
EPR spectrum of the solution was measured again.
Equipment settings: microwave power, 10 mW;
modulation amplitude, 1.0 G; time constant, 82 ms; scan
rate, 1.2 G/s; 4 scans per measurement.
Cyclic voltammograms were measured by using an Autolab
PGSTAT30 potentiostat and
a conventional three-
electrode cell consisting of a gold working electrode, a
platinum wire as the auxiliary electrode, and a reference
electrode made of Ag/AgNO3 (10 mM in acetonitrile;
+0.503 V vs. NHE). Tetrabutylammonium perchlorate
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 13
(Bu4NClO4, 0.10 M) was used as the electrolyte. UV-Vis resultant green solid in THF/MeOH for several times until
1
2
3
4
5
6
7
8
absorption spectroelectrochemistry was performed by
using a custom-built electrochemical cell consisting of a
gold minigrid working electrode, the Ag/AgNO3 (10 mM)
reference electrode used for cyclic voltammetry, and a
platinum wire auxiliary electrode mounted inside a quartz
cuvette with a path length of 25 μm. The application of
potentials was controlled by an EG&G PAR 173
potentiostat, and the spectra were recorded with a HP
8453A spectrophotometer.
the supernatant become colorless. The powder was dried
in an oven at 60oC for 24 hours prior to characterization.
Yield: 10.8 mg, 17%. MALDI-TOF-TOF MS (m/z): Calcd. for
1
[M+H]+: 1089.25. Found: 1089.34. H NMR in THF-d8 (,
ppm): 8.37 (dd, J = 9 Hz, 16H), 7.38 (t, J = 9 Hz, 16H).
Synthesis of ClPz. Same synthetic and purification
procedure as described for FPz, using 102 mg (0.34 mmol)
of the chloro-substituted dinitrile. Yield: 10.8 mg, 10%.
MALDI-TOF-TOF MS (m/z): Calcd. for [M+H]+: 1221.02.
9
1
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
The structure of the Pzs, unsubstituted Pz and phenyl
groups in this work was investigated with Time
Dependent-Density Functional Theory (TDDFT) as
implemented in the Gaussian 09 package,51 using B3LYP
and the 6-311G(d,p) basis set. After optimization of the
structures, the HOMO (Highest Occupied Molecular
Orbital) and LUMOs (Lowest Unoccupied Molecular
Orbital) were calculated.
Found: 1221.06. H NMR in THF-d8 (, ppm): 8.32 (d, J = 8
Hz, 16H), 7.64 (d, J = 8 Hz, 16H).
Synthesis of BrPz. Same synthetic and purification
procedure as described for FPz, using 60 mg (0.15 mmol) of
the bromo-substituted dinitrile. Yield: 6.0 mg, 10%.
MALDI-TOF-TOF MS (m/z): Calcd. for [M+H]+: 1576.61.
1
Found: 1576.71. H NMR in THF-d8 (, ppm): 8.25 (d, J = 8
Hz, 16H), 7.80 (d, J = 8 Hz, 16H).
Synthesis of the dinitriles. The well-known procedure
reported by Linstead describes the use of 2 eq. of Na for 1
eq. of the acetonitrile for the formation of dinitriles.52
However, in this work, the addition of a 10% excess of Na
in the reaction resulted in more reproducible yields, since
oxidation of this metal by air and moisture may easily
occur during handling. All dinitriles were obtained as a
mixture of cis and trans configurations and no attempt was
made to separate the isomers. The detailed synthetic
procedure and characterization of these precursors are
presented in the SI section.
Synthesis of MeOPz. Same synthetic and purification
procedure as described for FPz, using 75 mg (0.26 mmol)
of the methoxy-substituted dinitrile. Yield: 42 mg, 52%.
MALDI-TOF-TOF MS (m/z): Calcd. for [M+H]+: 1186.42.
Found: 1186.49. 1H NMR in THF-d8 (, ppm): 8.37 (d, J = 9
Hz, 16H), 7.15 (d, J = 9 Hz, 16H), 3.99 (s, 24H).
ASSOCIATED CONTENT
Supporting Information. Synthetic procedure for the
1
porphyrazines precursors; H NMR, 13C NMR and MALDI-
TOF-TOF spectra of all compounds; UV-vis, fluorescence and
singlet oxygen emission spectra for the porphyrazines in
solution; HOMO and LUMO calculations; flash photolysis and
EPR spectra for CF3Pz. “This material is available free of
charge via the Internet at http://pubs.acs.org.”
Synthesis of CF3Pz. Magnesium turnings (5.10 mg, 0.21
mmol) and 1-butanol (3 mL) were refluxed until complete
reaction of all magnesium (ca. 3 hours). The dinitrile (75
mg, 0.21 mmol) was added to the mixture and it was
refluxed for 12 hours under magnetic stirring and covered
from light. In the first hours the solution turns to dark
green, however, a longer time is needed to consume most
part of the dinitrile. The starting material is not fully
consumed during reaction because the trans isomer
cannot cyclize to form the porphyrazine ring. After cooling
to room temperature, the mixture was filtered under
vacuum and washed with methanol. The filtrate was
collected, the solvent was removed and the crude product
AUTHOR INFORMATION
Corresponding Authors
*baptista@iq.usp.br
*thiago@qui.ufmg.br
Author Contributions
All authors have given approval to the final version of the
manuscript.
was purified on
a silica gel column eluted with
Notes
cyclohexane/CH2Cl2 (1:2). The dark green solid was dried
in an oven at 60oC for 24 hours. Yield: 24.1 mg, 31%.
MALDI-TOF-TOF MS (m/z): Calcd. for [M+H]+: 1489.23.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
1
Found: 1489.28. H NMR in CDCl3 (, ppm): 8.41 (d, J = 8
The authors wish to thank Prof. Ohara Augusto and Dr.
Edlaine Linares for the EPR measurements, Prof. Daniel
Cardoso and Dr. Fernando Mattiucci for the transient
absorption measurements and Prof. Frank H. Quina for
revising the manuscript. We also thank FAPESP (scholarship
grant No 2016/15916-9 and research grants 2012/50680-5 and
2017/23416-9), CEPID-Redoxoma, NAP-Phototech-USP, CNPq
and CAPES (Finance Code 001) for the financial support.
Hz, 8H), 8.31 (s, 8H), 7.81 (d, J = 8 Hz, 8H), 7.70 (t, J = 8 Hz,
8H).
Synthesis of FPz. Same synthetic procedure as described
for CF3Pz, using 64 mg (0.24 mmol) of the fluoro-
substituted dinitrile. Due to its very limited solubility in
most organic solvents, the purification of this
porphyrazine was achieved by first heating the crude
reaction mixture in methanol, followed by filtration and
methanol washings, and then recrystallization of the
REFERENCES
10
ACS Paragon Plus Environment

Page 11 of 13
Journal of the American Chemical Society
(1)
Bottari, G.; Trukhina, O.; Ince, M.; Torres, T. Towards
Influence of the Triplet Excited State on the Photobleaching
Artificial Photosynthesis: Supramolecular, Donor-Acceptor,
Porphyrin- and Phthalocyanine/Carbon Nanostructure
Ensembles. Coord. Chem. Rev. 2012, 256 (21–22), 2453–2477.
https://doi.org/10.1016/j.ccr.2012.03.011.
Wibmer, L.; Lourenço, L. M. O.; Roth, A.; Katsukis, G.;
Neves, M. G. P. M. S.; Cavaleiro, J. A. S.; Tomé, J. P. C.; Torres,
T.; Guldi, D. M. Decorating Graphene Nanosheets with
Electron Accepting Pyridyl-Phthalocyanines. Nanoscale
2015, 7 (13), 5674–5682. https://doi.org/10.1039/c4nr05719h.
Kim, C. K.; Adhikari, A.; Deisseroth, K. Integration of
Optogenetics with Complementary Methodologies in
Systems Neuroscience. Nat. Rev. Neurosci. 2017, 18 (4), 222–
235. https://doi.org/10.1038/nm.2451.A.
Henderson, B. W.; Dougherty, T. J. How Does Photodynamic
Therapy Work? Photochem. Photobiol. 1992, 55 (1), 145–157.
https://doi.org/10.1111/j.1751-1097.1992.tb04222.x.
Baptista, M. S.; Cadet, J.; Di Mascio, P.; Ghogare, A. A.; Greer,
A.; Hamblin, M. R.; Lorente, C.; Nunez, S. C.; Ribeiro, M. S.;
Thomas, A. H.; Vignoni, M.; Yoshimura, T. M. Type I and
Type II Photosensitized Oxidation Reactions: Guidelines and
Mechanistic Pathways. Photochem. Photobiol. 2017, 93 (4),
912–919. https://doi.org/10.1111/php.12716.
Abrahamse, H.; Hamblin, M. R. New Photosensitizers for
Photodynamic Therapy. Biochem. J. 2016, 473, 347–364.
https://doi.org/10.1042/BJ20150942.
Martins, W. K.; Santos, N. F.; Rocha, C. de S.; Bacellar, I. O.
L.; Tsubone, T. M.; Viotto, A. C.; Matsukuma, A. Y.; Abrantes,
A. B. d. P.; Siani, P.; Dias, L. G.; Baptista, M. S. Parallel
Damage in Mitochondria and Lysosomes Is an Efficient Way
to Photoinduce Cell Death. Autophagy 2018, 15 (0), 259–279.
https://doi.org/10.1080/15548627.2018.1515609.
Albani, B. A.; Peña, B.; Leed, N. A.; De Paula, N. A. B. G.;
Pavani, C.; Baptista, M. S.; Dunbar, K. R.; Turro, C. Marked
Improvement in Photoinduced Cell Death by a New Tris-
Heteroleptic Complex with Dual Action: Singlet Oxygen
Sensitization and Ligand Dissociation. J. Am. Chem. Soc.
2014, 136 (49), 17095–17101. https://doi.org/10.1021/ja508272h.
Vakrat-Haglili, Y.; Weiner, L.; Brumfeld, V.; Brandis, A.;
Salomon, Y.; McIlroy, B.; Wilson, B. C.; Pawlak, A.;
Kinetics of Fluorescein in Microscopy. Biophys. J. 1996, 70
1
2
3
4
5
6
7
8
9
(6),
3495(96)79866-1.
Baptista, M. S.; Indig, G. L. Effect of BSA Binding on
Photophysical and Photochemical Properties of
2959–2968.
https://doi.org/10.1016/S0006-
(15)
(16)
(2)
(3)
Triarylmethane Dyes. J. Phys. Chem. B 2002, 102 (23), 4678–
4688. https://doi.org/10.1021/jp981185n.
Bacellar, I.; Oliveira, M. C.; Dantas, L.; Costa, E.; Junqueira,
H. C.; Martins, W. K.; Durantini, A. M.; Cosa, G.; Di Mascio,
P.; Wainwright, M.; Miotto, R.; Cordeiro, R. M.; Miyamoto,
S.;
Baptista
M.
S.
Photosensitized
Membrane
Permeabilization Requires Contact-Dependent Reactions
between Photosensitizer and Lipids. J. Am. Chem. Soc. 2018,
140, 9606–9615. https://doi.org/10.1021/jacs.8b05014.
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)
(5)
(17)
(18)
Tasso, T. T.; Furuyama, T.; Mack, J.; Nyokong, T.; Kobayashi,
N. Synthesis and Photophysical Investigation of
Tetraazaporphyrin Substituted with Aggregation-Induced
Emission (AIE) Active Moieties. Eur. J. Inorg. Chem. 2015,
2015 (33), 5516–5522. https://doi.org/10.1002/ejic.201500726.
Szczolko, W.; Sobotta, L.; Fita, P.; Koczorowski, T.; Mikus,
M.; Gdaniec, M.; Orzechowska, A.; Burda, K.; Sobiak, S.;
Wierzchowski, M.; Mielcarek, J.; Tykarska, E.; Goslinski, T.
Synthesis, Characteristics and Photochemical Studies of
(6)
(7)
Novel
Dimethylpyrrol-1-Yl
Tetrahedron Lett.
Porphyrazines
Possessing
Dimethylamino
53 (16),
Peripheral
2,5-
and
2012,
Groups.
2040–2044.
https://doi.org/10.1016/j.tetlet.2012.02.010.
Kobayashi, N.; Ashida, T.; Hiroya, K.; Osa, T. Benzene or
Naphthalene Ring-Fused Mono-Substituted Type
Tetraazaoctaphenylporphyrin Zinc Complexes. Chem. Lett.
1992, 1567–1570.
Sobbi, A. K.; Wohrle, D.; Schlettweinb, D. Photochemical
Stability of Various Porphyrins in Solution and as Thin Film
Electrodes. J. Chem. Soc. Perkin Trans. 2 1993, No. 3, 481–488.
https://doi.org/10.1039/p29930000481.
Yogo, T.; Urano, Y.; Ishitsuka, Y.; Maniwa, F.; Nagano, T.
Highly Efficient and Photostable Photosensitizer Based on
BODIPY Chromophore. J. Am. Chem. Soc. 2005, 127 (35),
12162–12163. https://doi.org/10.1021/ja0528533.
Song, L.; Hennink, E. J.; Young, I. T.; Tanke, H. J.
Photobleaching Kinetics of Fluorescein in Quantitative
Fluorescence Microscopy. Biophys. J. 1995, 68 (6), 2588–
2600. https://doi.org/10.1016/S0006-3495(95)80442-X.
Georgakoudi, I.; Foster, T. H. Singlet Oxygen- versus
Nonsinglet Oxygen-Mediated Mechanisms of Sensitizer
Photobleaching and Their Effects on Photodynamic
Dosimetry. Photochem. Photobiol. 1998, 67 (6), 612–625.
https://doi.org/10.1111/j.1751-1097.1998.tb09102.x.
Toyama, M. M.; Franco, M.; Catalani, L. H.; Araki, K.; Toma,
H. E. Spectroelectrochemical and Photophysical Properties
of a (3,4-pyridyl) Porphyrazine Supermolecule Containing
Four [Ru(bipy)2Cl]+ Groups. J. Photochem. Photobiol. A
Chem. 1998, 118, 11–17.
(19)
(20)
(21)
(22)
(23)
(8)
(9)
Rozanowska,
M.;
Sarna,
T.;
Scherz,
A.
The
Microenvironment Effect on the Generation of Reactive
Oxygen Species by Pd-Bacteriopheophorbide. J. Am. Chem.
Soc.
2005,
127
(17),
6487–6497.
https://doi.org/10.1021/ja046210j.
(10)
Lameijer, L. N.; Ernst, D.; Hopkins, S. L.; Meijer, M. S.; Askes,
S. H. C.; Le Dévédec, S. E.; Bonnet, S. A Red-Light-Activated
Ruthenium-Caged NAMPT Inhibitor Remains Phototoxic in
Hypoxic Cancer Cells. Angew. Chem. Int. Ed. 2017, 56 (38),
11549–11553. https://doi.org/10.1002/anie.201703890.
Bonnett, R.; Martinez, G. Photobleaching of Sensitizers Used
in Photodynamic Therapy. Tetrahedron 2001, 57 (591), 9513–
9547. https://doi.org/10.1016/S0040-4020(01)00952-8.
(24)
(25)
(11)
(12)
Arnaut, L. G.; Pereira, M. M.; Da̧browski, J. M.; Silva, E. F. F.;
Schaberle, F. A.; Abreu, A. R.; Rocha, L. B.; Barsan, M. M.;
Urbańska, K.; Stochel, G.; Brett, C. M. A. Photodynamic
Therapy Efficacy Enhanced by Dynamics: The Role of Charge
Transfer and Photostability in the Selection of
Photosensitizers. Chem. Eur. J. 2014, 20 (18), 5346–5357.
https://doi.org/10.1002/chem.201304202.
Pereira, M. M.; Monteiro, C. J. P.; Simões, A. V. C.; Pinto, S.
M. A.; Abreu, A. R.; Sá, G. F. F.; Silva, E. F. F.; Rocha, L. B.;
Dbrowski, J. M.; Formosinho, S. J.; Simões, S.; Arnaut, L. G.
Synthesis and Photophysical Characterization of a Library of
Photostable Halogenated Bacteriochlorins: An Access to
near Infrared Chemistry. Tetrahedron 2010, 66 (49), 9545–
9551. https://doi.org/10.1016/j.tet.2010.09.106.
Hansch, C.; Leo, A.; Taft, R. W. A Survey of Hammett
Substituent Constants and Resonance and Field Parameters.
Chem.
Rev.
1991,
91
(2),
165–195.
https://doi.org/10.1002/chin.199139332.
(26)
(27)
Cataldo, F. A Revision of the Gutmann Donor Numbers of a
Series of Phosphoramides Including Tepa. Eur. Chem. Bull
2015, 4 (2), 92–97. https://doi.org/10.17628/ECB.2015.4.92.
Díaz-Torres, R.; Alvarez, S. Coordinating Ability of Anions
and Solvents towards Transition Metals and Lanthanides.
(13)
(14)
Dalton
Trans.
2011,
40
(40),
10742–10750.
https://doi.org/10.1039/c1dt11184a.
(28)
Sobotta, L.; Fita, P.; Szczolko, W.; Wrotynski, M.;
Wierzchowski, M.; Goslinski, T.; Mielcarek, J. Functional
Singlet Oxygen Generators Based on Porphyrazines with
Song, L.; Varma, C. A.; Verhoeven, J. W.; Tanke, H. J.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 13
Peripheral 2,5-Dimethylpyrrol-1-yl and Dimethylamino
Computational, and Photochemical Studies. J. Phys. Chem. B
Groups. J. Photochem. Photobiol. A Chem. 2013, 269, 9–16.
https://doi.org/10.1016/j.jphotochem.2013.06.018.
Yoshida, T.; Zhou, W.; Furuyama, T.; Leznoff, D. B.;
Kobayashi, N. An Extremely Air-Stable 19π Porphyrinoid. J.
2005, 109 (20), 10107–10114. https://doi.org/10.1021/jp050591l.
Krammer, B.; Plaetzer, K. ALA and Its Clinical Impact, from
Bench to Bedside. Photochem. Photobiol. Sci. 2008, 7 (3),
283–289. https://doi.org/10.1039/b712847a.
Atif, M.; Zellweger, M.; Wagnieres, G. Review of the Role
Played by the Photosensitizer’s Photobleaching during
Photodynamic Therapy. J. Optoelectron. Adv. Mater. 2016, 18
(April), 338–350.
Isak, S. J.; Eyring, E. M. Fluorescence Quantum Yield of
Cresyl Violet in Methanol and Water as a Function of
Concentration. J. Phys. Chem. 1992, 58 (15), 1738–1742.
https://doi.org/10.1021/j100183a045.
1
2
3
4
5
6
7
8
(42)
(43)
(29)
(30)
Am.
Chem.
Soc.
2015,
137
(29),
9258–9261.
https://doi.org/10.1021/jacs.5b05781.
Schröder, J.; Himmel, D.; Kratzert, D.; Radtke, V.; Richert, S.;
Weber, S.; Böttcher, T. Isolation of a Stable Pyridine Radical
Anion. Chem. Commun. 2019, 55 (9), 1322–1325.
https://doi.org/10.1039/c8cc09700c.
Valko, M.; Izakovic, M.; Mazur, M.; Christopher, J.; Telser, J.
Role of Oxygen Radicals in DNA Damage and Cancer
Incidence. Mol. Cell. Biochem. 2004, 266, 37–56.
Sasaki, K.; Kashimura, T.; Masahiro, O.; Ohsaki, Y.; Ohta, N.
Solvent Effect in the Electrochemical Reduction of P-
Quinones in Several Aprotic Solvents. J. Electrochem. Soc.
1990, 137 (8), 2437–2443. https://doi.org/10.1149/1.2086957.
Wasylenko, D. J.; Ganesamoorthy, C.; Henderson, M. A.;
Berlinguette, C. P. Unraveling the Roles of the Acid Medium,
(44)
9
(31)
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
(45)
(46)
A Guide to Recording Fluorescence Quantum Yields
(http://www.horiba.com/fileadmin/uploads/Scientific/Doc
uments/Fluorescence/quantumyieldstrad.pdf).
(32)
Wilkinson, F.; Helman, W. P.; Ross, A. B. Quantum Yields
for the Photosensitized Formation of the Lowest
Electronically Excited Singlet State of Molecular Oxygen in
Solution. J. Phys. Chem. Ref. Data 1993, 22, 113–262.
Bacellar, I. O. L.; Pavani, C.; Sales, E. M.; Itri, R.; Wainwright,
M.; Baptista, M. S. Membrane Damage Efficiency of
Phenothiazinium Photosensitizers. Photochem. Photobiol.
2014, 90 (4), 801–813. https://doi.org/10.1111/php.12264.
Kuhn, H. J.; Braslavsky, S. E.; Schmidt, R. Chemical
Actinometry. IUPAC2004_Actin_March-23-04, 1-47.
Rodrigues, T.; De França, L. P.; Kawai, C.; De Faria, P. A.;
Mugnol, K. C. U.; Braga, F. M.; Tersariol, I. L. S.; Smaili, S. S.;
Nantes, I. L. Protective Role of Mitochondrial Unsaturated
Lipids on the Preservation of the Apoptotic Ability of
Cytochrome c Exposed to Singlet Oxygen. J. Biol. Chem.
(33)
(34)
Experimental
Probes,
and
Terminal
Oxidant,
(47)
(NH4)2[Ce(NO3)6], in the Study of a Homogeneous Water
Oxidation Catalyst. Inorg. Chem. 2011, 50 (8), 3662–3672.
https://doi.org/10.1021/ic2000188.
Shimizu, D.; Oh, J.; Furukawa, K.; Kim, D.; Osuka, A.
Triarylporphyrin Meso-Oxy Radicals: Remarkable Chemical
Stabilities and Oxidation to Oxophlorin π‑Cations. J. Am.
(48)
(49)
Chem.
Soc
2015,
137,
15584–15594.
https://doi.org/10.1021/jacs.5b11223.
(35)
(36)
Hannun, Y. A.; Obeid, L. M. Principles of Bioactive Lipid
Signalling: Lessons from Sphingolipids. Nat. Rev. 2008, 9,
139–150. https://doi.org/10.1038/nrm2329.
Di Mascio, P.; Martinez, G. R.; Miyamoto, S.; Ronsein, G. E.;
Medeiros, M. H. G.; Cadet, J. Singlet Molecular Oxygen
Reactions with Nucleic Acids, Lipids, and Proteins. Chem.
2007,
282
(35),
25577–25587.
https://doi.org/10.1074/jbc.M700009200.
(50)
(51)
Bacellar, I. O. L.; Baptista, M. S.; Junqueira, H. C.;
Wainwright, M.; Thalmann, F.; Marques, C. M.; Schroder, A.
P. Permeability of DOPC Bilayers under Photoinduced
Oxidation: Sensitivity to Photosensitizer. Biochim. Biophys.
Rev.
2019,
119
(3),
2043–2086.
https://doi.org/10.1021/acs.chemrev.8b00554.
Tsubone, T. M.; Junqueira, H. C.; Baptista, M. S.; Itri, R.
Contrasting Roles of Oxidized Lipids in Modulating
(37)
(38)
Acta
Biomembr.
2018,
1860
(11),
2366–2373.
https://doi.org/10.1016/j.bbamem.2018.06.001.
Membrane Microdomains. Biochim. Biophys. Acta
Biomembr. 2019, 1861 (3), 660–669.
https://doi.org/10.1016/j.bbamem.2018.12.017.
-
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.;
Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.;
Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.;
Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Vreven, T.; J. A. Montgomery, Jr.; Peralta, J. E.;
Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K.
N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox,
J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian,
Inc., Wallingford CT 2013.
Izquierdo, M. A.; Vysniauskas, A.; Lermontova, S. A.;
Grigoryev, I. S.; Shilyagina, N. Y.; Balalaeva, I. V.; Klapshina,
L. G.; Kuimova, M. K. Dual Use of Porphyrazines as
Sensitizers and Viscosity Markers in Photodynamic Therapy.
J. Mater. Chem.
B
2015,
3
(6),
1089–1096.
https://doi.org/10.1039/c4tb01678e.
(39)
(40)
(41)
Weber, G.; Charitat, T.; Baptista, M. S.; Uchoa, A. F.; Pavani,
C.; Junqueira, H. C.; Guo, Y.; Baulin, V. A.; Itri, R.; Marques,
C. M.; Schroder, A.P. Lipid Oxidation Induces Structural
Changes in Biomimetic Membranes. Soft Matter 2014, 10
(24), 4241–4247. https://doi.org/10.1039/c3sm52740a.
Ough, E.; Nyokong, T.; Stillman, M. J.; Creber, K. A. M.
Electrochemistry and Spectroscopy of Magnesium
Phthalocyanine. Analysis of the Absorption and Magnetic
Circular Dichroism Spectra. Inorg. Chem. 1988, 27 (15), 2724–
2732. https://doi.org/10.1021/ic00288a032.
(52)
Cook, A. H.; Linstead, R. P. Phthalocyanines. Part X I . The
D’Souza, F.; El-Khouly, M. E.; Gadde, S.; McCarty, A. L.; Karr,
P. A.; Zandler, M. E.; Araki, Y.; Ito, O. Self-Assembled via
Axial Coordination Magnesium Porphyrin-Imidazole
Appended Fullerene Dyad: Spectroscopic, Electrochemical,
Preparation
of
Octaphenylporphyrazines
from
Diphenylmaleinitrile. J. Chem. Soc. 1937, 929–933.
12
ACS Paragon Plus Environment

Page 13 of 13
Journal of the American Chemical Society
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
13
ACS Paragon Plus Environment