**3. DNA‐targeting porphyrin theranostics**

The abovementioned mechanisms of berberine and palmatine can be applied to porphyrin theranostics. For this purpose, cationic porphyrins are useful because they can be incorpo‐ rated into the cell nucleus and can photosensitize cellular DNA damage [53]. Furthermore, cationic porphyrins can bind to a DNA strand through electrostatic interaction, similar to ber‐ berine and palmatine. For example, anionic water‐soluble porphyrin PPIX hardly induces cel‐ lular and isolated DNA damage, whereas tetrakis(*N*‐methyl‐4‐pyridinio) porphyrin (TMPyP, see **Figure 8**) effectively photosensitizes the guanine‐specific oxidation of cellular and isolated DNA through 1 O2 generation. Thus, electron donor‐connecting cationic porphyrins were designed and synthesized to realize porphyrin theranostics.

#### **3.1. Binding interaction with DNA and cellular and isolated DNA‐damaging activity of water‐soluble porphyrins**

The effect of a DNA microenvironment on the photosensitized reaction of water‐soluble por‐ phyrin derivatives, TMPyP and its zinc complex (ZnTMPyP, see **Figure 8**), was reported [42]. The main driving force of DNA binding is electrostatic interaction. The binding form between these porphyrins and DNA depends on the concentration ratio of porphyrins and DNA bases. In the presence of a sufficient concentration of DNA, TMPyP mainly intercalates to the DNA strand, whereas ZnTMPyP binds to the DNA groove. An electrostatic interaction with DNA raises the redox potential of the binding porphyrins, resulting in suppression of the photoin‐ duced electron transfer from an electron donor to the DNA‐binding porphyrins, whereas the electron transfer from the porphyrins to the electron acceptor was enhanced.

Cellular DNA damage by photoirradiated water‐soluble porphyrins, TMPyP and PPIX was examined [53]. TMPyP and PPIX induced apoptosis in the human leukemia HL‐60 cell

**Figure 8.** Structures of TMPyP (left) and ZnTMPyP (right).

under photoirradiation [53]. TMPyP is incorporated in the cell nucleus and photosensi‐ tizes cellular DNA oxidation, whereas PPIX hardly demonstrates cellular DNA‐damaging ability. In the case of an isolated DNA fragment, photoexcited TMPyP effectively oxidized most guanine residues, whereas little or no DNA damage was observed in the PPIX case [53]. Consequently, a TMPyP cationic porphyrin should be useful as a DNA‐targeting photosensitizer.

#### **3.2. Design and synthesis of electron donor‐connecting porphyrin**

**3. DNA‐targeting porphyrin theranostics**

designed and synthesized to realize porphyrin theranostics.

DNA through 1

O2

176 Phthalocyanines and Some Current Applications

**Figure 8.** Structures of TMPyP (left) and ZnTMPyP (right).

**water‐soluble porphyrins**

The abovementioned mechanisms of berberine and palmatine can be applied to porphyrin theranostics. For this purpose, cationic porphyrins are useful because they can be incorpo‐ rated into the cell nucleus and can photosensitize cellular DNA damage [53]. Furthermore, cationic porphyrins can bind to a DNA strand through electrostatic interaction, similar to ber‐ berine and palmatine. For example, anionic water‐soluble porphyrin PPIX hardly induces cel‐ lular and isolated DNA damage, whereas tetrakis(*N*‐methyl‐4‐pyridinio) porphyrin (TMPyP, see **Figure 8**) effectively photosensitizes the guanine‐specific oxidation of cellular and isolated

**3.1. Binding interaction with DNA and cellular and isolated DNA‐damaging activity of** 

The effect of a DNA microenvironment on the photosensitized reaction of water‐soluble por‐ phyrin derivatives, TMPyP and its zinc complex (ZnTMPyP, see **Figure 8**), was reported [42]. The main driving force of DNA binding is electrostatic interaction. The binding form between these porphyrins and DNA depends on the concentration ratio of porphyrins and DNA bases. In the presence of a sufficient concentration of DNA, TMPyP mainly intercalates to the DNA strand, whereas ZnTMPyP binds to the DNA groove. An electrostatic interaction with DNA raises the redox potential of the binding porphyrins, resulting in suppression of the photoin‐ duced electron transfer from an electron donor to the DNA‐binding porphyrins, whereas the

Cellular DNA damage by photoirradiated water‐soluble porphyrins, TMPyP and PPIX was examined [53]. TMPyP and PPIX induced apoptosis in the human leukemia HL‐60 cell

electron transfer from the porphyrins to the electron acceptor was enhanced.

generation. Thus, electron donor‐connecting cationic porphyrins were

Molecular orbital (MO) calculation suggests that pyrene‐connecting TMPyP (PyTPP, see **Figure 9**) can be used for porphyrin theranostics in a DNA microenvironment [54]. **Figure 9** shows the optimized structures of PyTPP and AnTPP and their highest‐occu‐ pied MOs (HOMO). The binding action of PyTPP into the DNA major groove was sug‐ gested, and the apparent association constants, estimated from the relationship between the absorbance change and the DNA concentration, are relatively large (1.0 × 10<sup>6</sup> M−1 and 8.3 × 10<sup>5</sup> M−1 for AATT and AGTC, respectively). The fluorescence spectrum and its life‐ time measurements showed that this porphyrin demonstrates almost no fluorescence in

**Figure 9.** Structures of PyTPP (left) and AnTPP (right). The side‐view structures and the HOMO of these porphyrins were obtained by the MO calculation at the Hartree‐Fock 6‐31G\* level.

aqueous solution (Ф<sup>f</sup> < 0.001, see **Table 2**) because of the rapid intramolecular electron transfer. The electron‐accepting ability of the porphyrin moiety is decreased by the elec‐ trostatic interaction with DNA. In the presence of DNA, the fluorescence intensity was markedly increased (Ф<sup>f</sup> is 0.12 and 0.10 in the presence of 50‐μM base pairs AATT and AGTC, respectively). In addition, the typical near‐infrared emission spectrum of <sup>1</sup> O2 was clearly observed during the photoexcitation of PyTPP with DNA, whereas the emission was not observed without DNA. The estimated ФΔ by PyTPP‐DNA was 0.051 and 0.038 in the presence of 50‐μM base pairs AATT and AGTC, respectively. In conclusion, the S<sup>1</sup> state of PyTPP is effectively quenched by the pyrenyl moiety. The interaction with DNA sup‐ presses this electron transfer, leading to the enhancement of fluorescence emission. The intersystem crossing is also enhanced and makes 1 O2 generation possible.


The fluorescence properties and the ФΔ values were examined in a 10‐mM sodium phosphate buffer (pH = 7.6). These values were reported in the literature [54, 55].

**Table 2.** Fluorescence and photosensitized 1 O2 ‐generating activities of PyTPP and AnTPP in the absence or presence of DNA.

#### **3.3. Improvement of the activity control using anthracene**

In the abovementioned case of PyTPP, Ф<sup>f</sup> can be recovered to a value comparable to that of TMPyP. However, ФΔ is significantly smaller than that of TMPyP. A relatively small ФΔ value might be due to the self‐oxidation of PyTPP through the photosensitized 1 O2 genera‐ tion. Since an electron donor is easily oxidized by 1 O2 , the connection of the electron donor tends to decrease the apparent yield of 1 O2 generation. 1 O2 may oxidize the pyrene moiety through the Diels‐Alder reaction. To avoid this self‐oxidation, anthracene‐connecting TMPyP (AnTPP, see **Figure 9**) was designed and synthesized [55]. The optimized structure of AnTPP according to MO calculation suggested that oxidation of the anthracene moiety directly con‐ necting at the mesoposition of the porphyrin is difficult because of steric hindrance, resulting in recovery of the 1 O2 yield. In addition, the MO calculation indicated the steric rotational hindrance of the anthracene moiety around the mesoposition of the porphyrin, which keeps the two π‐electronic systems nearly orthogonal to each other. This calculation also showed that the activity control of fluorescence and <sup>1</sup> O2 generation of this porphyrin through an inter‐ action with DNA is possible.

In aqueous solution, AnTPP barely demonstrates fluorescence emission (Ф<sup>f</sup> < 0.001) and 1 O2 generation (**Table 2**). The observed fluorescence lifetime (<40 ps) indicates the rapid intramolecular electron transfer in the S1 state of the porphyrin moiety of AnTPP. AnTPP also binds to the DNA strand, mainly the minor groove, and the reported association con‐ stant is relatively large (~10<sup>6</sup> M−1). DNA‐binding AnTPP demonstrates a relatively strong fluorescence and long fluorescence lifetime comparable to those of the reference porphyrin without an electron donor. Furthermore, the 1 O2 ‐generating activity of AnTPP is recovered by DNA. The estimated values of ФΔ relative to that of methylene blue are 0.22 and 0.17 for the AATT‐ and AGTC‐binding forms of AnTPP, respectively (**Table 2**). The observed values of ФΔ are significantly larger than those of PyTPP. These results suggest that the 1 O2 ‐generating activity of AnTPP has improved due to the inhibition of self‐oxidation by the generated 1 O2 .

#### **3.4. Phenanthrene‐connecting cationic porphyrin**

aqueous solution (Ф<sup>f</sup>

178 Phthalocyanines and Some Current Applications

markedly increased (Ф<sup>f</sup>

intersystem crossing is also enhanced and makes 1

**Porphyrin DNA Ф<sup>f</sup>** *τ***<sup>f</sup>**

**3.3. Improvement of the activity control using anthracene**

tion. Since an electron donor is easily oxidized by 1

In the abovementioned case of PyTPP, Ф<sup>f</sup>

values were reported in the literature [54, 55].

**Table 2.** Fluorescence and photosensitized 1

of DNA.

tends to decrease the apparent yield of 1

O2

that the activity control of fluorescence and <sup>1</sup>

in recovery of the 1

action with DNA is possible.

< 0.001, see **Table 2**) because of the rapid intramolecular electron

is 0.12 and 0.10 in the presence of 50‐μM base pairs AATT and

generation possible.

**/ns (ratio) ФΔ**

can be recovered to a value comparable to that

‐generating activities of PyTPP and AnTPP in the absence or presence

generation of this porphyrin through an inter‐

, the connection of the electron donor

may oxidize the pyrene moiety

O2

genera‐

O2 was

state

transfer. The electron‐accepting ability of the porphyrin moiety is decreased by the elec‐ trostatic interaction with DNA. In the presence of DNA, the fluorescence intensity was

clearly observed during the photoexcitation of PyTPP with DNA, whereas the emission was not observed without DNA. The estimated ФΔ by PyTPP‐DNA was 0.051 and 0.038 in the presence of 50‐μM base pairs AATT and AGTC, respectively. In conclusion, the S<sup>1</sup>

of PyTPP is effectively quenched by the pyrenyl moiety. The interaction with DNA sup‐ presses this electron transfer, leading to the enhancement of fluorescence emission. The

> AATT 0.12 12.0 0.051 AGTC 0.10 10.6 (0.62) 2.8 (0.38) 0.038

AATT 0.098 10.4 (0.88) 3.6 (0.12) 0.22 AGTC 0.077 10.6 (0.79) 2.8 (0.21) 0.17

The fluorescence properties and the ФΔ values were examined in a 10‐mM sodium phosphate buffer (pH = 7.6). These

of TMPyP. However, ФΔ is significantly smaller than that of TMPyP. A relatively small ФΔ

through the Diels‐Alder reaction. To avoid this self‐oxidation, anthracene‐connecting TMPyP (AnTPP, see **Figure 9**) was designed and synthesized [55]. The optimized structure of AnTPP according to MO calculation suggested that oxidation of the anthracene moiety directly con‐ necting at the mesoposition of the porphyrin is difficult because of steric hindrance, resulting

hindrance of the anthracene moiety around the mesoposition of the porphyrin, which keeps the two π‐electronic systems nearly orthogonal to each other. This calculation also showed

O2

O2

O2

yield. In addition, the MO calculation indicated the steric rotational

generation. 1

value might be due to the self‐oxidation of PyTPP through the photosensitized 1

O2

O2

AGTC, respectively). In addition, the typical near‐infrared emission spectrum of <sup>1</sup>

PyTPP [54] Without <0.001 0.04 nd

AnTPP [55] Without <0.001 0.04 nd

O2

Phenanthrene was also used as the electron donor of the cationic porphyrin [56]. However, the activity control of the phenanthrene‐connecting porphyrin (PhenTPP, see **Figure 10**) was not successful. The MO calculation showed the HOMO location on the phenanthryl moiety of PhenTPP and predicted the similarity of this porphyrin property to the abovementioned PyTPP and AnTPP. However, the observed values of Φ<sup>f</sup> and *τ*<sup>f</sup> without DNA are 0.028 and 5.8 ns (89%) and 2.7 ns (11%), respectively, indicating insufficient quenching of the S<sup>1</sup> state by phenanthrene. Furthermore, the estimated value of ФΔ by PhenTPP without DNA is large (0.38). Consequently, the activity control of this type of porphyrin by phenanthrene is not appropriate. This result can be explained by the relatively small driving force of the intra‐ molecular electron transfer (−Δ*G* = 0.18 eV). The driving force dependence of this electron transfer is discussed in the next section in detail.

**Figure 10.** A structure of PhenTPP. The side‐view structure and the HOMO of PhenTPP (right) were obtained by the MO calculation at the Hartree‐Fock 6‐31G\* level.
