Photosensitized protein damage by dimethoxyphosphorus(V) tetraphenylporphyrin

Bioorganic & Medicinal Chemistry Letters 23 (2013) 2704–2707

Contents lists available at SciVerse ScienceDirect

Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl

Photosensitized protein damage by dimethoxyphosphorus(V) tetraphenylporphyrin Kazutaka Hirakawa a,⇑, Norihito Fukunaga a, Yoshinobu Nishimura b, Tatsuo Arai b, Shigetoshi Okazaki c a

Faculty of Engineering, Shizuoka University, Johoku 3-5-1, Naka-ku, Hamamatsu, Shizuoka 432 8561, Japan Department of Chemistry, University of Tsukuba, Tennodai 1-1-1, Tsukuba, Ibaraki 305 8571, Japan c Medical Photonics Research Center, Hamamatsu University School of Medicine, Handayama 1-20-1, Higashi-ku, Hamamatsu, Shizuoka 431 3192, Japan b

a r t i c l e

i n f o

Article history: Received 22 December 2012 Revised 14 February 2013 Accepted 18 February 2013 Available online 27 February 2013 Keywords: Phosphorus(V)porphyrin Photosensitizer Protein damage Electron transfer Singlet oxygen

a b s t r a c t For the purpose of the basic study of photodynamic therapy, the activity of the water-soluble P(V)porphyrin, dimethoxyP(V)tetraphenylporphyrin chloride (DMP(V)TPP), on photosensitized protein damage was examined. The quantum yield of singlet oxygen generation by DMP(V)TPP (0.64) was comparable with that of typical porphyrin photosensitizers. Absorption spectrum measurement demonstrated the binding interaction between DMP(V)TPP and human serum albumin, a water-soluble protein. Photo-irradiated DMP(V)TPP damaged the amino acid residue of human serum albumin, resulting in the decrease of the fluorescence intensity from the tryptophan residue of human serum albumin. A singlet oxygen quencher, sodium azide, could not completely inhibit the damage of human serum albumin, suggesting that the electron transfer mechanism contributes to protein damage as does singlet oxygen generation. The decrease of the fluorescence lifetime of DMP(V)TPP by human serum albumin supported the electron transfer mechanism. The estimated contribution of the electron transfer mechanism is 0.64. These results suggest that the activity of DMP(V)TPP can be preserved under lower oxygen concentration condition such as tumor. Ó 2013 Elsevier Ltd. All rights reserved.

Porphyrins are used as the drug for photodynamic therapy (PDT), which is a less invasive treatment of cancer and some non-malignant conditions.1–3 In general, administered photosensitizers damage cancer cells by the generation of singlet oxygen ð1 O2 Þ (Type II mechanism), which is formed through energy transfer to molecular oxygen from the photoexcited photosensitizer in cancer cell. However, the phototoxic effect of 1 O2 on the PDT is restricted because the oxygen concentration in a cancer cell is relatively low.4 Another important mechanism of photosensitized biomolecule damage is the oxidation reaction through electron transfer (Type I mechanism), which does not require oxygen.5 The electron transfer mechanism requires highly oxidative activity (a lower reduction potential) in the photoexcited state of the photosensitizer. Larger excitation energy is advantageous for the lower reduction potential of the photosensitizer in the photoexcited state. Ultra-violet photosensitizers mainly induce biomolecule photodamage through the electron transfer mechanism, whereas a visible-light photosensitizer is not appropriate for this mechanism. Therefore, it is important to select the appropriate molecular design to achieve electron transfer-mediated biomolecule damage using a visible-light photosensitizer. Since high-valent porphyrin complexes demonstrate a lower reduction potential in their ⇑ Corresponding author. Tel./fax: +81 53 478 1287. E-mail address: [email protected] (K. Hirakawa). 0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bmcl.2013.02.081

photoexcited state than free-base or low-valent metal complexes, these porphyrins are advantageous for the oxidative electron transfer reaction.5–12 Indeed, derivatives of high-valent porphyrin complexes, such as P(V)5,8 and Sb(V)12 complexes, photosensitize DNA damage through two mechanisms, that is, 1 O2 generation and the electron transfer reaction. In this study, photosensitized protein oxidation by a porphyrin P(V) complex, dimethoxyP(V)tetraphenylporphyrin chloride (DMP(V)TPP, Fig. 1) was examined. The specific characteristics of the porphyrin P(V) complexes are the variety of the substituted axial ligand and the relatively low redox potential of the one-electron reduction in the photoexcited state. In addition, P(V)porphyrin is cationic and water-soluble. DMP(V)TPP is the simplest molecule in the alkoxy-substituted P(V)porpyrins. As a target protein model, human serum albumin (HSA), a water-soluble protein, was used, because its structure and property were elucidated.13 DichloroP(V)tetraphenylporphyrin chloride (Cl2PP) was obtained by the phosphorus incorporation into commercially available tetraphenylporphyrin (Wako Chemicals Co., Osaka, Japan) according to the previous report.14 DMP(V)TPP was synthesized from Cl2PP by the previously reported method (Supplementary data). As a target biomacromolecule, HSA, a water-soluble protein, was used. The sample solution containing 10 lM DMP(V)TPP and 10 lM HSA in a sodium phosphate buffer (pH 7.6) was irradiated with a light-emitting diode (LED) (kmax = 519 nm, 1 mW cm2,

2705

K. Hirakawa et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2704–2707

ClH 3 CO

N +N P N N

OCH 3

Side View

Top View

Figure 1. Structure of DMP(V)TPP. The top view and side view were determined by the DFT calculation at B3LYP/6-31G⁄ level.

CCS Inc., Kyoto, Japan). The intensity of the LED light source was measured with a light meter (LM-331, AS ONE, Osaka, Japan). Protein damage by P(V)porphyrin was evaluated by measurement of the fluorescence intensity from the amino acid residues as previously reported.15 The excitation and detection wavelengths were 298 and 350 nm, respectively. Fluorescence decay of DMP(V)TPP was measured using a time-correlated single-photon counting method.16 In the presence of HSA, the hypochromic effect and red-shift were observed in the UV–vis absorption spectra of DMP(V)TPP (Fig. 2), indicating the static interaction between DMP(V)TPP and the protein. The intensity of HSA fluorescence around 350 nm, assigned to the tryptophan residue (Supplementary data), was decreased by photo-irradiation in the presence of DMP(V)TPP (Fig. 3). The fluorescence decrement of HSA can be explained by the amino acid oxidation through the photosensitized reaction.15 The quantum yield of tryptophan degradation photosensitized by DMP(V)TPP for 60 min irradiation was estimated from the decrease of the tryptophan fluorescence and the absorbed photon number by the porphyrin. The

Absorbance

2

+ HSA / μM 10 5 2 1 0.5 without

1.5 1 0.5 0 350

400

450

500

550

estimated yield was 1.8  105. The quantum yield of HSA photodamage by DMP(V)TPP was larger than that of tetrakis(1-methyl4-pyridinio)porphyrin (H2TMPyP) (1.2  105), which is a typical water soluble porphyrin photosensitizer.17 This HSA damage was partially inhibited by sodium azide, a physical quencher of 1 O2 (Fig. 3).18 Furthermore, HSA damage was enhanced in D2O (Supplementary data), in which the lifetime of 1 O2 is markedly elongated (about 2–4 ls in H2O to 70 ls in D2O).19 These findings suggest HSA oxidation by 1 O2 . The photosensitized 1 O2 generation by DMP(V)TPP was confirmed by the detection of near-infrared emission around 1270 nm (Fig. 4),20 which is assigned to the 1 O2 ð1 Dg Þ—3 O2 ð3 R g Þ transition. The quantum yield of 1 O2 generation (UD), which was estimated from the comparison of the emission intensity with that of methylene blue (0.52 in H2O),21 was 0.64. The relatively large value of UD indicate that the 1 O2 mechanism is also important for photosensitized biomolecule damage in the presence of a sufficient concentration of molecular oxygen. The 1 O2 emission was decreased by the addition of HSA. Although 1 O2 is quenched through the physical mechanism by protein not only the chemical reaction (protein oxidation), this result supported the HSA damage through 1 O2 mechanism. HSA damage was not completely inhibited by an excess amount of sodium azide (10 mM) (Fig. 3), suggesting that the electron transfer mechanism is partly responsible for HSA photodamage, as is the 1 O2 mechanism. The quenching rate coefficient of 1 O2 by sodium azide (kq) is almost diffusion control limit (kdif), which is calculated as follows:

kq  kdif ¼

600

650

ð1Þ

where R is the gas constant, T is the absolute temperature, and g is the viscosity of water (8.91  104 kg m1 s1). The quenching efficiency of 1 O2 by sodium azide (Efq) can be calculated from the following equation using the lifetime of 1 O2 (sD = 3.5 ls):

Efq ¼

Wavelength / nm

8000RT 3g

kq ½NaN3  kq ½NaN3  þ 1=sD

ð2Þ

where [NaN3] is the concentration of sodium azide. In the presence of 10 mM sodium azide, the Efq becomes 0.996. This value suggests that the almost all 1 O2 can be quenched by 10 mM sodium azide. Consequently, the damage of HSA photosensitized by DMP(V)TPP with 10 mM sodium azide should be due to the electron transfer mechanism. The inhibited ratio of the HSA damage by sodium azide indicates the contribution of the 1 O2 mechanism. The estimated contributions of the HSA damage through the electron transfer

DMP(V)TPP

Intensity (arb. unit)

[non-damaged HSA] / μM

Figure 2. Absorption spectra of DMP(V)TPP in the presence of HSA. The sample solution contained 10 lM DMP(V)TPP and HSA (1, 0.5, 1, 2, 5, or 10 lM) in a 10 mM sodium phosphate buffer (pH 7.6).

10

8 [NaN3] 10 mM 1 mM without

6

Methylene blue

DMP(V)TPP + HSA

4 0

20

40

60

80

100

120

Irradiation time / min Figure 3. Time course of HSA damage photosensitized by DMP(V)TPP and the effect of sodium azide (NaN3) on this HSA damage. The sample solution contained 10 lM DMP(V)TPP, 10 lM HSA, and indicated concentration of NaN3 in a 10 mM sodium phosphate buffer (pH 7.6).

1200

1250

1300

1350

1400

Wavelength / nm Figure 4. Near-infrared emission spectra of 1 O2 generated by the photosensitization of DMP(V)TPP or methylene blue. The sample solution contained 10 lM photosensitizer with or without 10 lM HSA in a 10 mM sodium phosphate buffer (pH 7.6).

2706

K. Hirakawa et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2704–2707

and the 1 O2 generation mechanisms for 60 min irradiation were 0.64 and 0.36, respectively. The free energy change (DG) for the electron transfer oxidation of the tryptophan residue by the photoexcited P(V)porphyrin was roughly calculated from the following equation:22

DG ¼ E0—0  eðEox  Ered Þ

ð3Þ

where E0–0 is the 0–0 transition energy of DMP(V)TPP (2.03 eV), e is the electronic charge, Eox is the oxidation potential of tryptophan (0.65 V vs SCE under the similar conditions of this study),23 and Ered is the reduction potential of DMP(V)TPP (0.50 V vs SCE in acetonitrile).6 Because the charge of the P(V)porphyrin is neutralized by the electron transfer, the factor of the distance between the electron donor and acceptor is negligible.7 The estimated values of DG (0.88 eV) suggest that the oxidation of the tryptophan residue of HSA through the electron transfer by the photoexcited DMP(V)TPP (the S1 state of DMP(V)TPP) is possible (Fig. 5). The damage of amino acids photosensitized by DMP(V)TPP was also examined using tryptophan, tyrosine, and phenylalanine with an HPLC. Tryptophan was clearly decomposed by the photoexcited DMP(V)TPP, whereas tyrosine and phenylalanine were hardly damaged by the photoexcited DMP(V)TPP. The time-resolved fluorescence intensity of DMP(V)TPP could be fitted by a single exponential function in the case without HSA, and the estimated lifetime (sf) was 4.8 ns. On the other hand, double exponential function was well fitted in the case with HSA, indicating that the microenvironment of DMP(V)TPP is affected through the interaction with HSA. The obtained values of sf were 4.7 ns (fraction: 61%) and 1.1 ns (39%). The longer lifetime is almost the same as that without HSA, suggesting that this component is not quenched by HSA. The shorter lifetime can be explained by the electron transfer quenching. These results suggest that DMP(V)TPP binds to HSA and the one microenvironment of DMP(V)TPP is in the vicinity of electron-donating amino acid, such as tryptophan residue. The transient fluorescence spectrum of the short lifetime species was quite similar to that of the longer lifetime species (Supplementary data), suggesting that the interaction between DMP(V)TPP and HSA is not strong. The quenching rate constant (ket) of photoexcited DMP(V)TPP through electron transfer in the HSA microenvironment should be estimated from the following equation:

ket ¼

crossing

T1 hν

1O 2

Energy Transfer

Oxidation of tryptophan

Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.02.081. References

6. 7. 8. 9.

12.

Type I mechanism S1

Electron Transfer CT

Decomposition of tryptophan

13. 14. 15. 16.

hν Reverse Electron Transfer

S0 Figure 5. Proposed mechanism of the tryptophan oxidation of HSA photosensitized by DMP(V)TPP. CT indicates the charge transfer state (DMP(V)TPP radical and tryptophan cation radical pair).

ð4Þ

This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government (Grant number 23750186).

11.

S0

1

Acknowledgments

10.

3O 2



s s0

where s is 1.1 ns and s0 is 4.8 ns. The obtained value of ket is 7.0  108 s1. This value suggests that the electron transfer from the amino acid residue to the S1 state of DMP(V)TPP reasonably proceeds. In conclusion, DMP(V)TPP could induce protein photodamage through 1 O2 generation and the electron transfer mechanism. 1 O2 generation is a well-known mechanism for porphyrin photosensitization (Fig. 5).24,25 The electron transfer mechanism is hardly observed in the case of protein or DNA damage by a visible-light photosensitizer.5 The time-resolved fluorescence study suggests that the electron abstraction from the tryptophan residue to the S1 of DMP(V)TPP contributes to the electron transfer mechanism of HSA photodamage. The formed radical cation of the tryptophan residue through electron transfer should undergo further reaction with the surrounding elements, such as water and oxygen. An oxidized product, such as N-formylkynurenine, should be finally formed.26 Singlet oxygen generated by HSA-binding DMP(V)TPP also oxidizes tryptophan residue, resulting in the formation of oxidized products including N-formylkynurenine. The total quantum yield of the protein photodamage by DMP(V)TPP was larger than that by H2TMPyP, a typical water-soluble porphyrin. The activity of DMP(V)TPP may be preserved under lower oxygen condition such as tumor.4 These character of DMP(V)TPP is advantageous for the photosensitizer of PDT.

1. 2. 3. 4. 5.

Type II mechanism Intersystem S1

1

17. 18. 19. 20.

21.

Dolmans, D. E. J. G. J.; Fukumura, D.; Jain, R. K. Nat. Rev. Cancer 2003, 3, 380. Castano, A. P.; Mroz, P.; Hamblin, M. R. Nat. Rev. Cancer 2006, 6, 535. Wilson, B. C.; Patterson, M. S. Phys. Med. Biol. 2008, 53, R61. Bratasz, A.; Kulkarni, A. C.; Kuppusamy, P. Biophys. J. 2007, 92, 2918. Hirakawa, K. In New Research on DNA Damage; Kimura, H., Suzuki, A., Eds.; Nova Science Publishers: New York, 2008; pp 197–219. Hirakawa, K.; Segawa, H. J. Photochem. Photobiol. A: Chem. 1999, 123, 67. Hirakawa, K.; Segawa, H. Photochem. Photobiol. Sci. 2010, 9, 704. Hirakawa, K.; Kawanishi, S.; Hirano, T.; Segawa, H. J. Photochem. Photobiol. B: Biol. 2007, 87, 209. Shiragami, T.; Matsumoto, J.; Inoue, H.; Yasuda, M. J. Photochem. Photobiol. C: Photochemistry Rev. 2005, 6, 227. Inoue, H.; Okamoto, T.; Kameo, Y.; Sumitani, M.; Fujiwara, A.; Ishibashi, D.; Hida, M. J. Chem. Soc., Perkin Trans. 1994, 1, 105. Andou, Y.; Ishikawa, K.; Shima, K.; Shiragami, T.; Yasuda, M.; Inoue, H. Bull. Chem. Soc. Jpn. 2002, 75, 1757. Hirakawa, K.; Kawanishi, S.; Matsumoto, J.; Shiragami, T.; Yasuda, T. J. Photochem. Photobiol. B: Biol. 2006, 82, 37. He, X. M.; Carter, D. C. Nature 1992, 358, 209. Marrese, C. A.; Carrano, C. J. Inorg. Chem. 1983, 1858, 22. Hirakawa, K.; Ebara, Y.; Hirano, T.; Segawa, H. J. Jpn. Soc. Laser Surg. Med. 2009, 29, 372. Hirakawa, K.; Hirano, T.; Nishimura, Y.; Arai, T.; Nosaka, Y. J. Phys. Chem. B 2012, 116, 3037. Carlos, L.; Pedersen, B. W.; Ogilby, P. R.; Mártire, D. O. Photochem. Photobiol. Sci. 2009, 10, 1080. Sima, J. Coord. Chem. Rev. 2006, 250, 2325. Ogilby, P. R.; Foote, C. S. J. Am. Chem. Soc. 1983, 105, 3423. Matsumoto, J.; Shinbara, T.; Tanimura, S.; Matsumoto, T.; Shiragami, T.; Yokoi, H.; Nosaka, Y.; Okazaki, S.; Hirakawa, K.; Yasuda, M. J. Photochem. Photobiol. A: Chem. 2011, 218, 178. Usui, Y.; Kamogawa, K. Photochem. Photobiol. 1974, 19, 245.

K. Hirakawa et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2704–2707 22. Weller, A. Z. Phys. Chem. (Noue Folge) 1982, 133, 93. 23. Nana, C. G.; Feng, Z. Z.; Li, W. X.; Ping, D. J.; Qin, C. H. Anal. Chim. Acta 2002, 452, 245. 24. DeRosa, M. C.; Crutchley, R. J. Coord. Chem. Rev. 2002, 233–234, 351.

2707

25. Lang, K.; Mosinger, J.; Wagneroviá, D. M. Coord. Chem. Rev. 2004, 248, 321. 26. Ehrenshaft, M.; Silva, S. O.; Perdivara, I.; Bilski, P.; Sik, R. H.; Chignell, C. F.; Tomer, K. B.; Mason, R. P. Free Radical Biol. Med. 2009, 46, 1260.