Synthesis and characterization of novel purpurinimides as photosensitizers for photodynamic therapy.

A series of novel purpurinimides with long wavelength absorption were designed and synthesized to develop novel and potential photosensitizers. These compounds were investigated through reduction, oxidation, rearrangement reaction and amidation reactions of methyl pheophorbide a. They demonstrated a considerable bathochromic shift of the major absorption band in the red region of the optical spectrum (695-704 nm). Newly synthesized purpurinimides were screened for their antitumor activities, and showed higher photodynamic efficiency against A549 cell lines as compared to purpurin-18 methyl ester. The results revealed the novel purpurinimides could be potential photosensitizers.

Introduction

Photodynamic therapy (PDT) is a relatively new n class="Disease">cancer modality that uses light, photosensitizers (PSs), and oxygen for the treatment of various forms of cancer by photodynamic action [1-3]. The PDT treatment derives great promise from the dual selectivity that is produced by both a preferential uptake of the drug by the diseased tissue and the restriction of carefully regulated light absorption onto the specific sites [4]. Regulatory approvals for the clinical use of PSs and PDT now exist in many countries around the world for treating cancers of the head and neck, brain, lung, pancreas, intraperitoneal cavity, breast, prostate and skin [5-7].

Chlorophyll (a natural dye) derivatives and related systems have been used as PSs in n class="Disease">cancer phototherapy. Chlorophylls exhibit photophysical properties similar to porphyrin systems; the Qy bands of chlorins is normally red-shifted to 20–30 nm and has a 10 times greater absorption intensity compared to porphyrins, which make chlorin-containing systems better candidates for PDT [8].

Purpurinimides, derived from chlorophyll-a, are tumor-avid PSs and show a strong absorption in the near IR region with a high singlet oxygen (1O2) producing efficiency, a key cytotoxic agent in PDT application [9]. PSs with long wavelength absorption should exhibit deeper tissue penetration, which is very useful to treat large and deeply seated tumors [10-12]. Therefore, the synthesis of novel purpurinimides has become the focus in PDT studies.

For chlorophyll derivatives, the Qy bands were strongly affected by introduced different substituent groups to Qy axis (N21–N23, see Scheme 1). In continuation with our earlier efforts on PS design [13], it was thought worthwhile to synthesize new n class="Chemical">purpurinimide derivatives by incorporating the essential structural features of the above-mentioned potential cytotoxic drugs in order to obtain synergistic effects, which we report herein.

Scheme 1.

Synthesis of purpurin-18-N-aminoimides. Reagents and conditions. (a) 5% H2SO4/methanol, room temperature (rt); (b) (i) Zn(OAc)2/methanol, (ii) H2, Pd/C, tetrahydrofuran (THF), (iii) trifluoroacetic acid (TFA); (c) KOH/1-propanol/air, rt; (d) corresponding amine, toluene, reflux; and (e) Ti(NO3)3/methanol, rt.

In the present work, we report the synthesis, structural characterization and biological evaluation of a series of novel purpurinimides through reduction, oxidation and rearrangement reactions followed by amidation reaction of n class="Chemical">methyl pheophorbide a (MPa).

Results and Discussion

Synthesis and Characterization

In our quest to compare the effect of imide analogs for quantitative structure-activity relationship (QSAR) studies [14], several attempts were made to convert the 3-vinyl group into the ethyl group and dimethyloxyethyl group by reduction and oxidation reaction.

The synthetic strategies adopted for the target compounds are shown in Scheme 1. MPa 1, as an important starting material, was isolated from n class="Chemical">chlorophyll paste (Excrementum bombycis). The vinyl group at 3-position of MPa 1 was selectively hydrogenated by using Pd/C as a catalyst to yield meso-pheophorbide a 2, which was converted to mesopurpurin-18 methyl ester 3 via air oxidation. Reaction of MPa 1 with Ti(NO3)3 at 0 °C produced the intermediate adduct 5 in high yield. Following the same procedure, 3-(2,2-dimethoxyethyl)-3-devinyl-purpurin-18 methyl ester 6 was obtained in 40% yield. From the reaction between two key intermediates 3 and 6, and the corresponding amines, (a) N,N-dimethyl ethylamine; (b) N,N-diethyl ethylamine; (c) N-isopropyl ethylamine; (d) N,N-dimethylpropyl ethylamine; and (e) imidazolyl propylamine, successfully afforded the final purpurinimides 4a–4e and 7a–7e in excellent yield.

All the purpurinimide compounds were successfully characterized by a con class="Chemical">mbination analysis of 1H NMR and UV-vis spectroscopies, and elemental analysis. The structures of these novel purpurinimides were confirmed by 1H NMR spectroscopy (Figure 1). Compared to the mesopurpurin-18 methyl ester 3 the 1H NMR spectra of 4a–4e showed each triplet at δ 4.54, 4.72, 4.58, 4.66 and 4.56 ppm for the protons of CO–N–CH2–, respectively. The imidazole protons of compound 4e have three singlet signals at δ 7.73, 7.15 and 7.10 ppm. In compounds 4a and 4d, N–(CH3)2 protons appear as a singlet at δ 2.73 and 1.99 ppm, respectively. And compound 4b shows a triplet at δ 1.27 ppm for the protons of N-(CH2CH3)2. Along with the proton signals in the chlorin macrocycle, signals of 3-(2,2-dimethoxyethyl) protons for 3-(2,2-dimethoxyethyl)-3-devinyl-purpurinimides were shown in δ 3.46 or 3.47 as a singlet. The 1H NMR spectra of 7a–7e showed each triplet at δ 4.55, 4.74, 4.37, 4.84 and 4.58 ppm for the protons of CO–N–CH2–, respectively. The imidazole protons of compound 7e reveal three singlet signals at δ 7.73, 7.16 and 7.09 ppm. In compounds 7a, N–(CH3)2 protons appear as a singlet at δ 2.78 ppm. Purpurinimide 7b shows a triplet at δ 1.44 ppm for the protons of N–(CH2CH3)2.

Figure 1.

The comparative 1H NMR spectra (CDCl3, 500 Hz) in the region of δ 4.0–10.0 ppm of purpurinimides 4a, 4e, 7a and 7e.

The spectroscopic properties of the purpurinimides in n class="Chemical">dichloromethane are shown in Figure 2 and summarized in Table 1. In the electronic absorption spectra, the long wavelength bands (Qy band) of mesopurpurinimides 4a–4e were observed in the range of 694–698 nm, compared with mesopurpurin-18 methyl ester 3 (686.7 nm), these mesopurpurinimides show a bathochromic shift of the Qy band. The electronic spectrum of 3-(2,2-dimethoxyethyl)-3-devinyl-purpurinimides shows a bathochromic shift of the Qy bands from 690.3 (compound 6) to 700.1 (compound 7a), 699.6 (compound 7b), 703.6 (compound 7c), 698.9 (compound 7d) and 698.8 nm (compound 7e). These purpurinimides had the “ideal” photochemical properties required for an effective PDT agent.

Figure 2.

Electronic absorption spectra of purpurinimides for (a) 3, 4a–4e; and (b) 6, 7a–7e in CH2Cl2.

Table 1.

Absorption properties of the purpurinimides (3, 4a–4e, 6, 7a–7e) in CH2Cl2.

Compound	Absorption λmax (nm) (log ɛ) a
	Soret	ΔSoret (Δɛ)	Qy	ΔQy (Δɛ)
3	410.8 (0.91)	0	686.7 (0.31)	0
4a	417.3 (0.92)	6.5 (0.01)	695.8 (0.29)	9.1 (−0.02)
4b	417.4 (0.93)	6.6 (0.02)	696.4 (0.28)	9.7 (−0.03)
4c	418.3 (0.91)	7.5 (0)	697.5 (0.27)	10.8 (−0.04)
4d	417.3 (0.91)	6.5 (0)	694.9 (0.26)	8.2 (−0.05)
4e	417.4 (0.92)	6.6 (0.01)	695.2 (0.25)	8.5 (−0.06)
6	411.5 (1.16)	0	690.3 (0.33)	0
7a	417.4 (1.19)	5.9 (0.03)	700.1 (0.36)	9.8 (0.03)
7b	417.2 (1.19)	5.7 (0.03)	699.6 (0.31)	9.3 (−0.02)
7c	418.1 (1.21)	6.6 (0.05)	703.6 (0.35)	13.3 (0.02)
7d	417.5 (1.21)	6.0 (0.05)	698.9 (0.35)	8.6 (0.02)
7e	417.4 (1.17)	5.9 (0.01)	698.8 (0.37)	8.5 (0.04)

ΔSoret, ΔQy and Δɛ represent the change of the Soret band, Qy band and absorbance intensity, respectively, between the substituted purpurinimides and corresponding starting materials.

Fluorescence (emission) spectra of purpurinimides 4a–4c and 7a–7c in n class="Chemical">dimethylsulfoxide (DMSO) are shown in Figure 3. For all measurements the excitation wavelength was 530 nm. Three broad emission bands were observed (for 4a, 550–644, 645–692, and 693–800 nm) with their emission maxima at 718 (4a), 717 (4b), 604 (4c), 720 (7a), 719 (7b), and 657 (7c) nm.

Figure 3.

Fluorescence spectra of purpurinimides 4a–4c and 7a–7c (1.0 × 10−4 M) in dimethylsulfoxide (DMSO).

In addition, elemental analysis data for the purpurinimides reveal a good match between calculated and experimental values to the compounds, respectively.

In Vitro Study

In vitro activity of the purpurinimides 4a–4c and 7a–7c was evaluated against n class="CellLine">A549 cell lines at various drug doses (1–10 or 1–20 μM) by MTT (3-(4,5-dimethylthiazole-2-yl)-2,5-biphenyl tetrazolium bromide) assay at 12 h incubation after photoirradiation (670–710 nm, total light dose 2 J/cm2 for 15 min). Compounds 4a–4c showed no dark cytotoxicity until the highest concentration (in this case 10 μM, Figure 4a), otherwise, compounds 7a–7c presented dark cytotoxicity more than 10 μM (Figure 4b). In all the compounds, upon photoirradiation, the cell viability was decreased consistent with increased concentration of drug dose (Figure 4). After photoirradiation, among all the purpurinimides, 4a reveals the best photodynamic activity result (IC50 0.28 μM, Table 2). The photodynamic activity is relatively higher in the order of 4a > 4b > 4c > 7c > 7a > 7b. These results suggest that the relatively different photodynamic activity results are significantly dependent on the functional groups at 3-position as well as various amino moieties on the chlorin macrocycle. Consequently, the novel purpurinimides exhibit excellent photodynamic activity, showing potential PSs for PDT.

Figure 4.

Comparative in vitro dark cytotoxicity and phototoxicity results for 4a–4c and 7a–7c at various drug doses (1–20 μM) against A549 cell lines by MTT (3-(4,5-dimethylthiazole-2-yl)-2,5-biphenyl tetrazolium bromide) assay at 12 h incubation after photoirradiation (670–710 nm, total light dose 2 J/cm2 for 15 min); (a) 4a–4c; and (b) 7a–7c. The data are expressed as mean of three experiments.

Table 2.

IC50 values of the purpurinimides against A549 cell lines at 12 h incubation time after photoirradiation.

Compound	4a	4b	4c	7a	7b	7c
IC50 a (μM)	0.28	0.37	0.40	0.95	1.27	0.70

IC50 presents the half maximal (50%) inhibitory concentration of the compound.

Singlet Oxygen Study

Figure 5 reveals relative difference of 1O2 photogeneration between the n class="Chemical">purpurinimides 4a–4c and 7a–7c using 1,3-diphenylisobenzofuran (DPBF) as a selective 1O2 acceptor [15]. Among the purpurinimides, 4c and 7a showed relatively higher 1O2 photogeneration, results which are slightly different from the in vitro results. Consequently, this result proves that the increased photodynamic activity attributed to the purpurinimides induced 1O2 photogeneration through cellular penetration and localization of the purpurinimides into the cells.

Figure 5.

Comparative absorbance decay (%) of 1,3-diphenylisobenzofuran (DPBF) (50 μM in DMSO) at 418 nm after photoirradiation (total light dose 2 J/cm2, irradiation time 15 min) in the absence (control) and presence of 1 μM of MB (methylene blue), 4a–4c and 7a–7c. The data are expressed as mean of three experiments.

The Hydrophobicity Property Study

It is generally believed that the hydrophobicity parameter (logarithm of the partition coefficient between n-octanol and n class="Chemical">water; log P) is closely related to the cellular uptake. We applied a program module of the ACD/Labs software (version 12.01, Advanced Chemistry Development, Inc., Toronto, ON, Canada) to calculate the lipophilicity of the synthesized purpurinimides. Results are summarized in Table 3. All the purpurinimides have log P values up to log 5, which theoretically should be expected to give good PDT tumor response.

Table 3.

Hydrophobicity parameters (log P) of the purpurinimides calculated by means of computer software ACD/Labs (version 12.01).

Compound	Log P
4a	6.94 ± 1.62
4b	8.00 ± 1.62
4c	7.32 ± 1.62
4d	6.89 ± 1.63
4e	6.88 ± 1.62
7a	5.80 ± 1.64
7b	6.87 ± 1.64
7c	6.19 ± 1.64
7d	5.76 ± 1.65
7e	5.75 ± 1.64

Experimental Section

General Methods

The 1H NMR spectra were recorded on a Varian-500 MHz spectrometer (Varian, Palo Alto, CA, USA). Chemical shifts are given as δ values using TMS as the internal standard and J values in Hz. The UV-visible spectra were recorded on S-3100 spectrophotometer (Scinco, Seoul, Korea) using n class="Chemical">dichloromethane as solvent. Elemental analysis was performed on Flash 2000 series of automatic elemental analyzer (Thermo Fisher Scientific, Milano, Italy) at Biohealth Products Research Center (BPRC), Inje University, Korea. Fluorescence spectra were obtained using a LS-50B Perkin Elmer luminescence spectrometer (Perkin Elmer, Waltham, MA, USA) at the Center for Research Facilities, Gyeongsang National University, Korea. The hydrophobicity parameter (logarithm of the partition coefficient between n-octanol and water; log P) was calculated on the basis of the purpurinimide structure using ACD/Labs software (version 12.01). Melting points (uncorrected) were measured on an Electrothermal IA9000s Series digital melting point apparatus. Thin-layer chromatography (TLC) was done on Merck silica gel 60 glass sheets (Cat. HX948839, layer thickness 0.25 mm) (Merck, Darmstadt, Germany). Column chromatography was performed over silica gel 60 (230–400 mesh) (Merck). In some cases, preparative TLC plates were also used for the purification (Analtech precoated silica gel GF glass plate, Cat. 01012, layer thickness 0.5 mm) (Merck). Materials obtained from commercial suppliers were used without further purification. MPa 1 [16], mesopurpurin-18 methyl ester 3 [9,16] and 3-(2,2-dimethoxyethyl)-3-devinyl-purpurin-18 methyl ester 6 [9,17] were prepared according to the literature procedures.

Preparation of Mesopurpurin-18-N-aminoimides

General Procedure

In a typical experiment, mesopurpurin-18 methyl ester 3 (200 mg) and excess of each corresponding n class="Chemical">amine (0.15 mL) were dissolved in toluene (20 mL), and the mixture was refluxed under nitrogen atmosphere. After TLC showed complete consumption of purpurin-18 methyl ester, the mixture was cooled to room temperature, and then the solvent and excess amines were removed. The crude product was purified using silica column chromatography or preparative TLC plates with 10% methanol in dichloromethane to give corresponding purpurinimide 4a, 4b, 4c, 4d and 4e as a purple solid, respectively.

Characteristic Data for Mesopurpurin-18-N-(N,N-dimethyl)ethylimide 4a

Yield: 96%. Mp: 99–101 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.3 (0.92), 476.8 (0.04), 508.0 (0.05), 545.7 (0.15), 695.8 (0.29). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.49 (s, 1H, 10H), 9.13 (s, 1H, 5H), 8.48 (s, 1H, 20H), 5.32 (m, 1H 17H), 4.72 (t, J = 7.0 Hz, 2H, N–CH–CH2–N–(CH3)2), 4.33 (q, J = 7.5 Hz, 1H, 18H), 3.75 (s, 3H, 121CH3), 3.72 (m, 2H, 31CH2), 3.58 (s, 3H, 172CO2CH3), 3.51 (m, 2H, 81CH2), 3.22 (s, 3H, 21CH3), 3.13 (s, 3H, 71CH3), 2.73 (s, 6H, N–(CH3)2), 2.70, 2.42 and 1.96 (m, 6H, N–CH–CH2–N–(CH3)2, 2 × 171H and 2 × 172H), 1.76 (d, J = 7.0 Hz, 3H, 181CH3), 1.68 (t, J = 7.5 Hz, 3H, 82CH3), 1.63 (t, J = 7.5 Hz, 3H, 32CH3), 0.10 and −0.08 (each br s, 1H, 2NH). Anal. calcd. for C38H46N6O4: C, 70.13; H, 7.12; N, 12.91. Found: C, 70.18; H, 7.15; N, 12.92.

Characteristic Data for Mesopurpurin-18-N-(N,N-diethyl)ethylimide 4b

Yield: 95%. Mp: 118–120 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.4 (0.93), 477.1 (0.04), 507.6 (0.06), 546.0 (0.15), 649.6 (0.07), 696.4 (0.28). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.47 (s, 1H, 10H), 9.11 (s, 1H, 5H), 8.49 (s, 1H, 20H), 5.34 (m, 1H 17H), 4.58 (m, 2H, N–CH–CH2–N–(CH3)2), 4.33 (q, J = 7.5 Hz, 1H, 18H), 3.74 (s, 3H, 121CH3), 3.71 (q, J = 7.5 Hz, 2H, 31CH2), 3.57 (s, 3H, 172CO2CH3), 3.50 (m, 2H, 81CH2), 3.22 (s, 3H, 21CH3), 3.11 (s, 3H, 71CH3), 2.89 (q, J = 7.0 Hz, 4H, N–(CHCH3)2), 2.70, 2.40, 2.35 and 2.00 (m, 6H, N–CH2–CH–N–(CH2CH3)2, 2 × 171H and 2 × 172H), 1.75 (d, J = 7.5 Hz, 3H, 181CH3), 1.67 (t, J = 7.5 Hz, 3H, 82CH3), 1.61 (t, J = 7.5 Hz, 3H, 32CH3), 1.13 (t, J = 6.0 Hz, 6H, N–(CH2–CH)2), 0.01 and −0.17 (each br s, 1H, 2NH). Anal. calcd. for C40H50N6O4: C, 70.77; H, 7.42; N, 12.38. Found: C, 70.80; H, 7.44; N, 12.39.

Characteristic Data for Mesopurpurin-18-N-(N-isopropylamino)ethylimide 4c

Yield: 96%. Mp: 100–102 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 418.3 (0.91), 477.0 (0.04), 508.2 (0.06), 5497.5 (0.16), 650.9 (0.08), 697.5 (0.27). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.30 (s, 1H, 10H), 9.04 (s, 1H, 5H), 8.47 (s, 1H, 20H), 5.32 (m, 1H 17H), 4.66 (m, 2H, N–CH–CH2–NH–), 4.36 (q, J = 7.5 Hz, 1H, 18H), 3.68 (q, J = 7.5 Hz, 2H, 31CH2), 3.64 (s, 3H, 121CH3), 3.57 (s, 3H, 172CO2CH3), 3.47 (m, 2H, 81CH2), 3.21 (s, 3H, 21CH3), 3.17 (m, J = 7.5 Hz, 1H, NH–CH–(CH3)2), 3.05 (s, 3H, 71CH3), 2.71, 2.40, 2.06 and 1.97 (m, 6H, N–CH2–CH–NH–, 2 × 171H and 2 × 172H), 1.75 (d, J = 7.5 Hz, 3H, 181CH3), 1.66 (t, J = 7.5 Hz, 3H, 82CH3), 1.56 (t, J = 7.5 Hz, 3H, 32CH3), 1.18 (d, J = 6.0 Hz, 6H, NH–CH–(CH)2), −0.03 and −0.17 (each br s, 1H, 2NH). Anal. calcd. for C39H48N6O4: C, 70.46; H, 7.28; N, 12.64. Found: C, 70.50; H, 7.29; N, 12.67.

Characteristic Data for Mesopurpurin-18-N-(N,N-dimethylpropylamino)propylimide 4d

Yield: 97%. Mp: 105–107 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.3 (0.91), 476.8 (0.04), 507.6 (0.06), 544.8 (0.13), 638.0 (0.05), 694.9 (0.26). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.44 (s, 1H, 10H), 9.10 (s, 1H, 5H), 8.47 (s, 1H, 20H), 5.26 (m, 1H 17H), 4.56 (m, 4H, N–CH–CH2–CH–), 4.32 (q, J = 7.5 Hz, 1H, 18H), 3.71 (s, 3H, 121CH3), 3.70 (m, 2H, 31CH2), 3.54 (s, 3H, 172CO2CH3), 3.49 (m, 2H, 81CH2), 3.22 (s, 3H, 21CH3), 3.11 (s, 3H, 71CH3), 2.95 (t, J = 7.0 Hz, 2H, N–CH2–CH2–CH2–NH–CH2–CH2–CH–N–(CH3)), 2.72–2.64, 2.52–2.28, 2.21–2.15 (m, 8H, N–CH2–CH2–CH2–NH–CH–CH–CH2–N–(CH3), 2 × 171H and 2 × 172H), 1.99 (s, 6H, N–(CH3)2), 1.75 (d, J = 7.0 Hz, 3H, 181CH3), 1.68 (t, J = 7.5 Hz, 3H, 82CH3), 1.64 (m, 2H, N–CH2–CH–CH2–NH–CH2–CH–CH2–N–(CH3)), 1.62 (t, J = 7.0 Hz, 3H, 32CH3), 0.20, 0.08 (each br s, 1H, 2NH). Anal. calcd. for C42H55N7O4: C, 69.88; H, 7.68; N, 13.58. Found: C, 69.92; H, 7.71; N, 13.56.

Characteristic Data for Mesopurpurin-18-N-(imidazolyl)propylimide 4e

Yield: 96%. Mp: 84–86 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.4 (0.92), 478.1 (0.04), 508.0 (0.06), 545.3 (0.13), 638.2 (0.06), 695.2 (0.25). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.36 (s, 1H, 10H), 9.06 (s, 1H, 5H), 8.48 (s, 1H, 20H), 7.73, 7.15, 7.10 (s, 3H, imidazole–H), 5.31 (m, 1H 17H), 4.34 (q, J = 7.0 Hz, 1H, 18H), 4.54, 4.27 (each m, 4H, N–CH–CH2–CH–), 3.69 (q, J = 7.5 Hz, 2H, 31CH2), 3.68 (s, 3H, 121CH3), 3.54 (s, 3H, 172CO2CH3), 3.48 (m, 2H, 81CH2), 3.21 (s, 3H, 21CH3), 3.06 (s, 3H, 71CH3), 2.71, 2.50, 2.36 and 2.02 (m, 6H, N–CH2–CH–CH2–, 2 × 171H and 2 × 172H), 1.78 (d, J = 7.5 Hz, 3H, 181CH3), 1.67 (t, J = 7.5 Hz, 3H, 82CH3), 1.59 (t, J = 7.5 Hz, 3H, 32CH3), 0.02 and −0.13 (each br s, 1H, 2NH). Anal. calcd. for C40H45N7O4: C, 69.85; H, 6.59; N, 14.25. Found: C, 69.88; H, 6.61; N, 14.26.

Preparation of 3-(2,2-Dimethoxyethyl)-3-devinyl-purpurin-18-N-aminoimides

In a typical experiment, 3-(2,2-dimethoxyethyl)-3-devinyl-purpurin-18 methyl ester 6 (200 mg) and excess of each corresponding n class="Chemical">amine (0.15 mL) were dissolved in toluene (20 mL), and the mixture was refluxed under nitrogen atmosphere. After TLC showed complete consumption of purpurin-18 methyl ester, the mixture was cooled to room temperature, and then the solvent and excess amines were removed. The crude product was purified using silica column chromatography or preparative TLC plates with 10% methanol in dichloromethane to give corresponding purpurinimide 7a, 7b, 7c, 7d and 7e as a purple solid, respectively.

Characteristic Data for 3-(2,2-Dimethoxyethyl)-3-devinyl-purpurin-18-N-(N,N-dimethyl)ethyl-imide 7a

Yield: 96%. Mp: 10–103 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.4 (1.19), 509.3 (0.06), 546.9 (0.18), 648.1 (0.08), 700.1 (0.36). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.54 (s, 1H, 10H), 9.26 (s, 1H, 5H), 8.52 (s, 1H, 20H), 5.33 (m, 1H 17H), 4.95 (t, J = 5.5 Hz, 1H, 32H), 4.74 (t, J = 7.0 Hz, 2H, N–CH–CH2–N–(CH3)2), 4.34 (q, J = 7.5 Hz, 1H, 18H), 4.02 (d, J = 5.0 Hz, 2H, 31CH2), 3.76 (s, 3H, 121CH3), 3.62 (q, J = 8.0 Hz, 2H, 81CH2), 3.58 (s, 3H, 172CO2CH3), 3.46 (s, 6H, 32(OCH3)2), 3.27 (s, 3H, 21CH3), 3.15 (s, 3H, 71CH3), 2.78 (s, 6H, N–(CH3)2), 2.70, 2.40 and 1.97 (m, 6H, N–CH2–CH–N–(CH3)2, 2 × 171H and 2 × 172H), 1.77 (d, J = 7.5 Hz, 3H, 181CH3), 1.65 (t, J = 7.5 Hz, 3H, 82CH3), 0.04 and −0.08 (each br s, 1H, 2NH). Anal. calcd. for C40H50N6O6: C, 67.58; H, 7.09; N, 11.82. Found: C, 67.60; H, 7.11; N, 11.85.

Characteristic Data for 3-(2,2-Dimethoxyethyl)-3-devinyl-purpurin-18-N-(N,N-diethyl)ethyl-imide 7b

Yield: 96%. Mp: 105–107 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.2 (1.19), 509.2 (0.06), 546.5 (0.16), 649.2 (0.08), 699.6 (0.31). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.50 (s, 1H, 10H), 9.24 (s, 1H, 5H), 8.52 (s, 1H, 20H), 5.27 (m, 1H, 17H), 4.94 (t, J = 5.5 Hz, 1H, 32H)), 4.37 (3, 3H, N–CH–CH2–N–(CH2CH3) and 18H), 4.01 (d, J = 5.5 Hz, 2H, 31CH2), 3.73 (s, 3H, 121CH3), 3.61 (q, J = 8.0 Hz, 2H, 81CH2), 3.57 (s, 3H, 172CO2CH3), 3.47 (s, 6H, 32(OCH3)2), 3.27 (s, 3H, 21CH3), 3.14 (s, 3H, 71CH3), 2.90 (q, J = 7.0 Hz, 4H, N–(CHCH3)2), 2.71, 2.40 and 2.00 (m, 6H, N–CH2–CH–N–(CH2CH3)2, 2 × 171H and 2 × 172H), 1.81 (d, J = 7.0 Hz, 3H, 181CH3), 1.64 (t, J = 7.5 Hz, 3H, 82CH3), 1.44 (t, J = 6.0 Hz, 6H, N–(CH2–CH)2), 0.13 and 0.08 (each br s, 1H, 2NH). Anal. calcd. for C42H54N6O6: C, 68.27; H, 7.37; N, 11.37. Found: C, 68.30; H, 7.39; N, 11.40.

Characteristic Data for 3-(2,2-Dimethoxyethyl)-3-devinyl-purpurin-18-N-(N-isopropylamino)-ethylimide 7c

Yield: 95%. Mp: 128–130 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 418.1 (1.21), 509.4 (0.07), 549.2 (0.21), 651.7 (0.14), 703.6 (0.35). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.07 (s, 1H, 10H), 8.78 (s, 1H, 5H), 8.48 (s, 1H, 20H), 5.21 (m, 1H 17H), 4.93 (t, J = 5.5 Hz, 1H, 32H), 4.84 (m, 2H, N–CH–CH2–NH–), 4.31 (q, J = 7.5 Hz, 1H, 18H), 4.00 (d, J = 5.5 Hz, 2H, 31CH2), 3.67 (q, J = 6.5 Hz, 2H, 81CH2), 3.55 (s, 3H, 172CO2CH3), 3.47 (s, 6H, 32(OCH3)2), 3.35 (m, J = 7.5 Hz, 1H, NH–CH–(CH3)2), 3.27 (s, 3H, 21CH3), 2.97 (s, 3H, 71CH3), 2.70, 2.37, 2.06 and 1.92 (m, 6H, N–CH2–CH–NH–, 2 × 171H and 2 × 172H), 1.82 (d, J = 7.0 Hz, 3H, 181CH3), 1.52 (d, J = 6.0 Hz, 6H, NH–CH–(CH)2), 1.41 (t, J = 7.5 Hz, 3H, 82CH3), 0.88 and −0.23 (each br s, 1H, 2NH). Anal. calcd. for C41H52N6O6: C, 67.93; H, 7.23; N, 11.59. Found: C, 69.97; H, 7.26; N, 11.62.

Characteristic Data for 3-(2,2-Dimethoxyethyl)-3-devinyl-purpurin-18-N-(N,N-dimethylpropyl-amino)propylimide 7d

Yield: 98%. Mp: 96–98 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.5 (1.21), 508.2 (0.07), 546.2 (0.18), 661.0 (0.09), 698.9 (0.35). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.37 (s, 1H, 10H), 9.15 (s, 1H, 5H), 8.48 (s, 1H, 20H), 5.27 (m, 1H 17H), 4.96 (t, J = 5.5 Hz, 1H, 32H), 4.91, 4.58 (each t, 4H, N–CH–CH2–CH–), 4.34 (q, 1H, 18H), 4.03 (d, J = 5.5 Hz, 2H, 31CH2), 3.65 (s, 3H, 121CH3), 3.55 (s, 3H, 172CO2CH3), 3.49 (m, 2H, 81CH2), 3.46 (s, 6H, 32(OCH3)2), 3.45 (s, 3H, 21CH3), 3.10 (s, 3H, 71CH3), 2.94 (t, J = 6.5 Hz, 2H, N–CH2–CH2–CH2–NH–CH2–CH2–CH–N–(CH3)), 2.72–2.65, 2.44–2.30, 2.21–2.14 (m, 8H, N–CH2–CH2–CH2–NH–CH–CH–CH2–N–(CH3), 2 × 171H and 2 × 172H), 1.99 (s, 6H, N–(CH3)2), 1.76 (d, J = 7.5 Hz, 3H, 181CH3), 1.66 (m, 2H, N–CH2–CH–CH2–NH–CH2–CH–CH2–N–(CH3)), 1.59 (t, J = 7.0 Hz, 3H, 82CH3), 0.04 (br s, 1H, NH). Anal. calcd. for C44H59N7O6: C, 67.58; H, 7.60; N, 12.54. Found: C, 67.61; H, 7.62; N, 12.55.

Characteristic Data for 3-(2,2-Dimethoxyethyl)-3-devinyl-purpurin-18-N-(imidazolyl)-propylimide 7e

Yield: 97%. Mp: 87–89 °C. UV-vis in CH2Cl2, λmax (nm, rel. intensity log ɛ), 417.4 (1.17), 509.1 (0.07), 545.9 (0.18), 645.0 (0.08), 698.8 (0.37). n class="Chemical">1H NMR (500 MHz, CDCl3): δ 9.51 (s, 1H, 10H), 9.24 (s, 1H, 5H), 8.53 (s, 1H, 20H), 7.16, 7.09 (s, 3H, imidazole–H), 5.34 (m, 1H 17H), 4.94 (t, J = 5.5 Hz, 1H, 32H), 4.55, 4.28 (each m, 4H, N–CH–CH2–CH–), 4.35 (q, J = 7.0 Hz, 1H, 18H), 4.00 (d, J = 5.5 Hz, 2H, 31CH2), 3.75 (s, 3H, 121CH3), 3.57 (q, J = 7.5 Hz, 2H, 81CH2), 3.54 (s, 3H, 172CO2CH3), 3.46 (s, 6H, 32(OCH3)2), 3.27 (s, 3H, 21CH3), 3.14 (s, 3H, 71CH3), 2.73, 2.54, 2.09 and 2.01 (m, 6H, N–CH2–CH–CH2–, 2 × 171H and 2 × 172H), 1.77 (d, J = 7.0 Hz, 3H, 181CH3), 1.64 (t, J = 7.5 Hz, 3H, 82CH3), 0.02 and −0.11 (each br s, 1H, 2NH). Anal. calcd. for C42H49N7O6: C, 67.45; H, 6.60; N, 13.11. Found: C, 67.48; H, 6.62; N, 13.13.

In Vitro Photosensitizing Efficacy

A549 cell lines were cultured at 37 °C in a humidified 5% n class="Chemical">CO2 incubator using RFMI 1640 growth medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. For phototoxicity studies, A549 cells were plated in 96-well plates at a density of 1 × 105 cells/well. After 24 h of incubation, 100 μL of 1, 2, 5, 10 and 15 μM purpurinimides were added in each well, respectively. Plates were returned to the incubator for 24 h. And then the cells were replaced with fresh media and exposed to light (640–710 nm, total dose 2.0 J/cm2 for 15 min). Following illumination, the plates were incubated at 37 °C in the dark. After 3, 12 and 24 h incubations, MTT solution was added into each well and the absorbance was measured by fluorescence multi-detection reader (BioTek, Synergy HT, Winooski, VT, USA) at 450 nm. Each group consisted of 3 wells. The percentage cell survival was calculated by normalization with respect to the value for no PS treatment (control).

Measurement of Singlet Oxygen Photogeneration

1,3-Diphenylisobenzofuran (n class="Chemical">DPBF) was used as a selective 1O2 acceptor, which was bleached upon reaction with 1O2 [15]. Five sample solutions of DPBF in DMSO (50 μM) containing, respectively, DPBF only (50 μM, control sample), DPBF + methylene blue (MB) (1 μM), DPBF + 4a (1 μM), DPBF + 4b (1 μM), DPBF + 4c (1 μM), DPBF + 7a (1 μM), DPBF + 7b (1 μM), DPBF + 7c (1 μM) were prepared in dark. All the samples were placed in a 96-well plate and the container was covered with aluminum foil. The samples were irradiated (2 J/cm2) for 15 min. After irradiation, visible spectra of the sample solutions were measured spectrophotometrically. The normalized absorbances of DPBF at 418 nm in these samples were reported. The 1O2 photogeneration activities of MB, 4a–4c, and 7a–7c can be compared with the different absorbance decay of each sample relative to the DPBF control sample.

Conclusions

We described the synthesis of novel purpurinimides, n class="Chemical">mesopurpurin-18-N-aminoimides and 3-(2,2-dimethoxyethyl)-3-devinyl-purpurin-18-N-aminoimides with various amines (N,N-dimethyl ethylamine, N,N-diethyl ethylamine, N-isopropyl ethylamine, N,N-dimethylpropyl ethylamine and imidazolyl propylamine). The final desired purpurinimides were obtained in excellent yield. The purpurinimides were characterized by a combination analysis of 1H NMR, UV-vis and photoluminescence spectroscopies, and elemental analysis. 1H NMR spectroscopy confirms the structures of purpurinimides using ammonium formation after methylation, resulting in significant down field shift of the methyl proton signals. In the electronic absorption spectra, compared with starting materials, these purpurinimides show a bathochromic shift (8–13 nm) of the Qy band, resulting in a long wavelength absorption (695–704 nm) which should be helpful for promoting deep light penetration into tumor tissue because of minimal light scattering. Fluorescence spectra present three broad emission bands at 550–800 nm range. Preliminary in vitro studies demonstrate that the new purpurinimides revealed excellent photodynamic efficacy (IC50 0.28–1.27 μM at 12 h incubation time after photoirradiation), which corresponds the excellent 1O2 photogeneration of the purpurinimides. Among the purpurinimides, mesopurpurin-18-N-(N,N-dimethyl)ethylimide 4a presents the best photodynamic activity result. The photodynamic activity is relatively higher in the order of 4a > 4b > 4c > 7c > 7a > 7b, results which are related to the hydrophobicity property (log P). For the purpurinimides to be potential candidates for PDT, further in vivo studies (e.g., pharmacokinetics and tissue distribution tests) are desirable, which is currently under investigation. This result could be useful for synthesis and development of new potential PSs as well as for understanding of QSAR study in PDT.