J. CHEM. SOC. FARADAY TRANS., 1994, 90(8), 1073-1076 Photophysical Studies of Substituted Porphyrins Paul Charlesworth and T. George Truscott* Chemistry Department, Keele University, Staffordshire, UK ST5 5BG David Kessel Pharmacology and Medicine, Wayne State University School of Medicine, Detroit, MI 4820 1, USA Craig J. Medforth and Kevin M. Smith Department of Chemistry, University of California, Davis, CA 95616, USA Ground-state, fluorescence and triplet-state parameters, together with singlet oxygen quantum yields are pre-sented for two porphyrins with extensive substituents at both the p-pyrrole and meso positions. The relevance of the results to the photodynamic therapy of cancer is discussed. Recently there has been considerable interest in the intro- duction of steric hindrance to porphyrins and phthalocya- nines and the effect which this has upon the photophysical properties of the molecule.' Deviations from planarity in photosynthetic chromophores such as the chlorophylls have been suggested to be pivotal in their behaviour, and in model porphyrins, steric hindrance between B and meso substituents has been shown to induce non-planar conformations.Theo- retical calculations based on some photosynthetic reaction centres have shown that changes in the conformation of the molecule would affect it in such a way as to change the highest occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals, leading to a change in their photo-chemical properties.'~~ It is also thought that steric hindrance may be the way forward to a reduction in porphyrin and phthalocyanine aggregation, which would be important if such molecules are to be used as sensitizers of singlet oxygen in the photodynamic therapy of cancer. Studies of two symmetrically substituted porphyrins having substituents at both the /3-pyrrole and meso positions are presented.Dodecaphenylporphyrin (DPP) [Fig. l(a)], has all the p-pyrrole and meso positions occupied by phenyl rings, and 2,3,7,8,12,13,17,18-octaethyl-5,10,15,20-tetraphenyl-OETPP Fig. 1 Structuresofdodecaphenylporphyrin,DPP and 2,3,7,8,12,13, 17,18-octaethyl-5,10,15,20-tetraphenylporphyrin,OETPP porphyrin (OETPP) [Fig. l(b)] has all the B-pyrrole positions occupied by ethyl groups and the meso positions occupied by phenyl rings.Under the conditions studied in this work both porphyrins exist entirely as monomers. However, we choose methanol and benzene as solvents to allow study of the cationic and neutral forms of the porphyrins, respectively. Previous work by Medforth and Smith concerned the con- formational structure of DPP4 and the zinc complex and free base of OETPP.4 By X-ray crystallography, these workers demonstrated' that both of the molecules adopt a non-planar saddle conformation with alternating pyrrole rings pointing up or down. They also showed, by variable-temperature proton NMR experiments, that the molecules are also non- planar in solution and may undergo a macrocyclic inversion process. Experimental DPP and OETPP were synthesized as described previo~sly~.~and used without further purification. All non- deuteriated solvents were of HPLC grade and were obtained from Aldrich Chemicals.Deuteriated methanol was obtained at 99.9% D from Goss Chemicals. Absorption spectra were measured using a PC-controlled Perkin-Elmer Lambda 2 spectrophotometer. Fluorescence spectra were measured using a PC-controlled Perkin-Elmer LS50 spectrofluorimeter. Fluorescence lifetime studies were made at the SERC Daresbury Synchrotron facility on station 12.1 (time-resolved spectros~opy).~*~ The time resolution of this facility is ca. 100 ps. Measurements were made in benzene and methanol; in all cases the absorbance at the excitation wavelength was <0.1 in order to reduce problems with inner filtering.The triplet studies were made, using nitrogen-saturated benzene and methanol solutions, by laser flash photolysis with 355 or 532 nm excitation from the frequency-tripled or frequency-doubled outputs of an Nd : YAG laser (Spectron Lasers, 16 ns pulse width). Triplet spectra presented are nor- malised to a laser energy of 1. The triplet-state molar absorp- tion coefficients were determined to +lo% by the complete conversion method, and intersystem crossing quantum yields were determined to f10% by the comparative method with anthracene in cyclohexane as the actinometer.'-' Singlet oxygen was monitored via its near-IR luminescence at 1270 nm due to the 02('Ag)+Oz(3Zi)radiative tran- sition, following 532 nm laser excitation from the frequency- doubled output of the Nd: YAG laser with a detection system comprising a germanium photodiode and Judson amplifier, linked via a Tektronix digital oscilloscope to a PC for data analysis.Singlet oxygen yields were measured to f10% relative to meso-tetraphenylporphine (Porphyrin Products) in benzene (djA = 0.63)," and haematoporphyrin (Porphyrin Products) in methanol (aA= 0.53)13as standards. Biological Studies Murine leukemia L1210 cells were grown in Fischer's medium (GIBCO, Grand Island, NY) supplemented with 10% horse serum and antibiotics. Incubations were carried out in growth medium with 20 mmol dm-3 HEPES pH 7.2 replacing sodium hydrogencarbonate. To introduce porphyrins into cells a 10 mmol dm-3 solu-tion in N,N'dimethylformamide was added to cell suspen-sions in the incubation medium (10% serum).The level of the solvent so used is not sufficient to affect any parameter being measured. Also, note the hydrophobicity of the drugs them-selves is not a problem with regard to clinical use. Cells were incubated with sensitizers for 30 min at 37 "C, collected by centrifugation, suspended in fresh medium at 10 "Cto minimize temperature-dependent repair systems, and irradiated (time = 3 min) using a 600 W QH lamp (5 mW cm-', 1800 J m-'). The wavelength of irradiation was limited to 630-850 nm by cut-off and heat-absorbing filters and further attenuated by a 10 cm layer of water. The resulting fluence rate was measured with a calibrated EG & G 450-1 radiometer; this was calculated to be 2.1 f0.15 mW cm-' per 10 nm bandwidth over the range 650-750 nm.Results and Discussion It is well established that the two porphyrins studied are strong bases. In methanol they exist as cations, while in benzene they are in the neutral form but can be readily con-verted to the cationic form. From the ground-state spectra (Fig. 2) it is clear that solvent plays an important role in the photophysical properties, and this is shown to some extent in many of the studies we have made on these molecules. The effect of changing solvent from benzene to methanol is to cause a loss of the Q bands in the region 500-800 nm which 1.o 0.8 0.6 , ,\ 0.4 \ 7 \ f\ \ I \ \ 0.2 \ L 0 J.CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 are replaced by one band at ca. 730 nm and accompanied by a shift in the wavelength and/or a broadening of the Soret band, this may be attributed to protonation of the porphyrin, as an effect which has been noted previously for OETPP' and DPP.4 The fluoresence maxima were observed for OETPP at ca. 758 nm in methanol, and ca. 755 nm in benzene; with those for DPP being at ca. 796 nm in methanol and ca. 780 nm in benzene. The low fluorescence quantum yields obtained for these compounds (OETPP: djf xO.005 in methanol and @jf x0.004 in benzene; DPP: @jf x0.002 in methanol and @jf x0.003 in benzene) correlate well with the time-resolved studies made at the Daresbury Laboratory.Lifetimes obtained for OETPP (Aex = 480 nm Aem = 760 nm) were 0.21 ns (91.5%),0.37 ns (8.5%) in methanol and 0.92 ns (ca. 100%) in benzene. With the lifetimes for DPP (Aex = 460 nm, A,, = 780 nm) being <0.1 ns (ca. 90%), 0.28 ns (ca. 10%) in meth-anol and 0.82 ns (ca. 100%) in benzene. The biexponential lifetimes in methanol are of the order of 100-400 ps, suggest-ing that protonation leads to enhanced deactivation of the energy, either through intra-or inter-molecular interactions, providing more than one possible pathway for the rapid energy loss. The reason for the minor (<10%) component in methanol is not clear since the porphyrins exist as the mono-mers in this solvent and mixtures of free base with the dica-tion would not seem to account for the data. The position of the triplet maximum for each of the indi-vidual compounds is largely unaffected by the choice of solvent in this study.For DPP (Fig. 3) in benzene the maximum is at ca. 500 nm (E~x72200 dm3 mol-I cm-') and in methanol is at ca. 510 nm (complete conversion not achieved). For OETPP (Fig. 4) the maximum is at ca. 550 nm (E~x21 OOO dm3 mol-' cm-')in benzene and at ca. 550 nm (complete conversion not achieved) in methanol. As with the singlet state lifetimes, the triplet state lifetimes are rather short (DPP, z x2.5 ps in methanol and 7 x20 ps in benzene; OETPP, z x2.8 ps in methanol and 7 x20 ps in benzene) and the triplet quantum yields also rather small (DPP, di, x 0.29; OETPP, @jT x 0.34 in benzene) compared to planar 2.0 2.0 (') ,-, 'I I I 1.5 I 1.o 1.o \ Y nrn I \4-'5 0.5 \ 0.5I e i 3 0.3- .0.3 x5 r I\ 0.2- I '- -0.2 I '\ z 0.10.1 --// \\ ' -0.1 A/nm 0 300 400 500 A/n m -, 600 0 700 800 Fig.2 Ground-state absorption spectra for (a)DPP in benzene (free base); (b)DPP in methanol (dication+free base); (c) OETPP in benzene (free base) and (d)OETPP in methanol (dication) J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 1075 0.10 (a 1 A 0.10 benzene. The value for S, (the fraction of triplets which when quenched by ground-state oxygen lead to singlet oxygen) was 0.051 iz 10.05 calculated according to the equation I This takes into account the short lifetimes of the triplet state16 and yield S, x 0.6, which is somewhat lower than normally expected for a porphyrin (ca.0.8). This not only provides an explanation for the lower @, values, but suggests that processes other than energy transfer occur in the excited- state oxygen complex. Whilst these yields are low relative to 0.2 their standards (TPP @, x 0.63 in benzene and HP.2HC1 @, x 0.53 in methanol) they are still sufficient as potential 0 sensitizers for photodynamic therapy. Also, as there is a red shift of the n-x* transition, this leads to a molecule with a lower singlet-state energy and hence a -0.2 -0.2 lower triplet-state energy than most porphyrins. Further- O.:: more, if the energy of the triplet state is very close to that of oxygen then the ability of the porphyrin to sensitize singlet oxygen formation may be reduced.It would seem unlikely that sterically hindered triplet energy transfer to the ground- state molecular oxygen would be responsible for the decrease 1-0.64 400 1-0.6 in the quantum yield because triplet energy transfer is largely 350 450 500 550 600 650 700 unaffected by steric hindrance, since energy transfer at van R/nm der Waals separation is so rapid.17 However, Scaiano et Fig. 3 Triplet minus singlet difference spectra for (a) DPP in all8 suggest that slower triplet energy transfer may be a func- benzene and (b)DPP in methanol tion of stereoelectronic factors rather than steric hindrance and may provide an indicator to the nature of the triplet porphyrins such as meso-tetraphenylporphine (z x 100 ps, state, be it n,x*, z,n*,a combination, or even that there are djf x 0.1 @= x 0.8).This may be attributed, as discussed by fewer accessible orientations allowing the orbital overlap and Tsuchiya" to intramolecular energy necessary for energy transfer. We found the rate of reaction of Takeda et ~1.'~ transfer, leading to energy dissipation as heat, or possibly OETPP and DPP triplet states with ground-state molecular intramolecular electron transfer between phenyl groups. oxygen was not significantly slower than one-ninth the Singlet oxygen yields obtained by time-resolved lumines- diffusion-controlled rate in benzene (DPP, k, x 1.6 x lo9 cence techniques are OETPP, @, x 0.25 in MeOD; @, x dm3 mol-' s-'; OETPP, k, x 1.2 x lo9 dm3 mol-' s-l in 0.24 in benzene and DPP, a, x 0.24 in MeOD; @, x 0.19 in benzene).However, in methanol (DPP, k, x 0.25 x lo9 dm3 benzene. The @, values are lower than the @= values in mol-' s-'; OETPP, kqx0.31x lo9 dm3 mol-' s-') we found the rate constant to be almost one order of magnitude 0.4 slower. We did, however, find that the relative yields of singlet oxygen were comparable or only mariginally higher 0.2 for both porphyrins in methanol than in benzene. Thus, despite the rather short triplet lifetimes in methanol, the 0 higher oxygen concentration in this solvent leads to substan- tial singlet oxygen production. -0.2 Biological Studies-0.4 Extracellular levels of the different sensitizers required to -0.6 photosensitize lethally 50% of a cell population (EC,,V) were determined.The corresponding cellular levels were estimated -0.8 from distribution ratios as described above. Viability was 0.2 assessed by the MTT assay" carried out after a 3 day incu- bation of cells in culture. This latter procedure provides infor- mation on photodamage which is well correlated with results Oa2i;'..-----I 0 0 of clongenic assays.20-2' For this assay, cells were diluted 50-fold with fresh medium. The concentration of photo-sensitizers was sufficiently reduced so that no 'dark' toxicity -0.2 was observed. Table 1 Cell kill data for murine leukemia L1210 cells with DPP and OETPP -0.41-JLJ-0-4350 400 450 500 550 600 650 700-0.6 -0.6 sensitizer EC,,/pmol dm-' IC,,/pmol dm-' DR A/nm DPP 30 40 1.3 OETPP 5 10 2.0 Fig.4 Triplet minus singlet difference spectra for (a) OETPP in PP 1 8 8.8 benzene and (b)OETPP in methanol EC,, (Table 1) is the extracellular drug level needed to photosensitize lethally 50% of a cell population, and the IC5, value represents the corresponding intracellular drug concen- tration, calculated from the distribution ratio (DR) which represents the ratio of intracellular/extracellular drug levels. Data represent the average of three determinations which dif- fered by < f10% of the numbers shown. Conclusion The similarities of the photophysical properties between the two molecules may be mirrored in their biological activities where DPP, although requiring a four-fold increase in intra- cellular level to yield the same phototoxic effect as OETPP, shows a level of activity almost identical to that of OETPP when the differences in absorption coefficient are taken into account.Furthermore, whilst the photophysics suggests that both drugs would be equally good at producing phototoxic products, DPP may localise at sites more resistant to cyto- toxic intermediates than the site of OETPP binding. Unfor- tunately, the fluorescence lifetimes are so short that it is not possible to identify the sites of localisation accurately from fluorescence microscopy. By comparison when a 1 pmol dm-3 aluminium phthalocyanine solution is used, giving a distribution ratio of about 60, hence an intracellular drug level of about 60 pmol dm-3, irradiation for 15 s is sufficient to kill 50% of the cells, which is an order of magnitude less than required by DPP for the same cell kill.Also, when a 1 pmol dm-3 concentration of protoporphyrin is used, resulting in a distribution ratio of 8, hence an intracellular drug level of 8 pmol dm-3, irradiation for 3 min is sufficient to reduce the viability by 50%. This is approximately the same value as was obtained with OETPP, indicating that non-planarity does not necessarily interfere with photo-dynamic efficacy. P.C. and T.G.T. acknowledge the Cancer Research Campaign (UK) for financial support and T.G.T.also thanks the E.E.C. P.D.T. EURONET (ERBCHRXCT930178) for financial support. C.J.M. thanks the Fulbright commission (Travel Scholarship) and the Associated Western Universities (Post- J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Doctoral Fellowship). K.M.S. acknowledges a grant from the National Science Foundation. (CHE-93-05577). D.K. acknowledges NIH grant CA 52997. References 1 B. D. Rihter, M. D. Bohorquez, M. A. J. Rodgers and M. E. Kenney, Photochem. Photobiol., 1992,55, 677. 2 K. M. Barkigia, D. M. Berber, J. Fajer, C. J. Medforth, M. W. Renner and K. M. Smith, J. Am. Chem. SOC., 1990,112,8851. 3 K. M. Barkigia, L. Chantranupong, K. M. Smith and J. Fajer, J. Am. Chem. SOC., 1988,110,7566. 4 C. J. Medforth and K.M. Smith, Tetrahedron Lett., 1990, 31, 5583. 5 C. J. Medforth, M. 0. Senge, K. M. Smith, L. D. Sparks and J. A. Shelnutt, J. Am. Chem. SOC., 1992,114,9859. 6 R. Sparrow, R. G. Brown, E. H. Evans and D. Shaw, J. Chem. SOC.,Faraday Trans. 2,1986,82,2249. 7 H. Winick, Sci. Am., 1987,257,88. 8 R. Bensasson and E. J. Land, Trans. Faraday SOC., 1971, 67, 1904. 9 R. V. Bensasson, E. A. Dawe and E. J. Land, J. Chem. SOC., Faraday Trans. I, 1977,73,1319. 10 I. Carmichael and G. L. Hug, J. Phys. Chem. Ref: Data, 1986,15, 1. 11 D. Lavelette, R. Bensasson, B. Amand and E. J. Land, Chem. Phys. Lett., 1971, 10, 331. 12 R. Bonnett, D. J. McGarvey, A. Harriman, E. J. Land, T. G. Truscott, and J. Winfield, Photochem. Photobiol., 1988,48,271. 13 R. W. Redmond, K. Heihoff, S. E. Braslavsky and T. G. Truscott, Photochem. Photobiol., 1987,45,209. 14 J. Takeda, 0.Toshie and M. Sato, Chem. Phys. Lett., 1991, 183, 384. 15 S. Tsuchiya, Chem. Phys. Lett., 1990, 169, 608. 16 C. Knox, E. J. Land and T. G. Truscott, J. Photochem. Photo- biol, B: Biology, 1988, 1, 315. 17 P. Wagner, J. M. McGrath and R. G. Zepp, J. Am. Chem. SOC., 1972,94,6883. 18 J. C. Scaiano, W. J. Leigh, M.A. Meador and P. J. Wagner, J. Am. Chem. SOC., 1985,107,5806. 19 R. A. Plumb, R. Milroy and S. B. Kaye, Cancer Res., 1989, 49, 4435. 20 A. P. McHale and L. McHale, Cancer Lett., 1988,41, 315. 21 D. Kessel, A. Morgan and G. M. Garbo, Photochem. Photobiol., 1991,54,193. Paper 3/05 169B;Received 26th August, 1993