Synthesis and characterization of photodynamic activity of an iodinated Chlorin p₆ copper complex

Abstract

We report the synthesis of a new iodinated Chlorin p₆ copper complex (ICp₆–Cu) and its efficacy for photodynamic treatment (PDT) of cancer cells. The metal complex is obtained by reacting Chlorin p₆ (Cp₆) with copper iodide (CuI). The complex formation results in a shift in the Q absorption band of Cp₆ from 663 nm to 634 nm and X-ray fluorescence of the complex showed the presence of both copper and iodine. FTIR and EPR spectroscopy suggests that the copper is attached to Cp₆ at the two adjacent carboxylic groups. Studies on the photochemical generation of singlet oxygen (¹O₂) and other reactive oxygen species (ROS) using fluorescence probes revealed that ICp₆–Cu acts predominantly through the type I process. PDT of oral cancer cells with ICp₆–Cu (10 μM, 3 h) and red light (630 ± 20 nm, ∼12 J cm⁻²) led to ∼90% phototoxicity. Furthermore, in contrast to Cp₆, the phototoxicity induced by ICp₆–Cu is not significantly affected under hypoxic conditions.

---

Introduction

Photodynamic therapy (PDT) of cancer utilizes a photosensitive drug referred to as a photosensitizer (PS) and irradiation with light of appropriate wavelengths to induce tumor damage via generation of ROS.^1,2 The absorption of light by the PS results in its excitation to a singlet excited state (¹PS*) which undergoes inter system crossing (ISC), to form a relatively long lived excited-triplet state (³PS). The ³PS can undergo two types of reactions defined as type I and type II process. In type I process, the ³PS reacts with surrounding substrates involving transfer of hydrogen atom or electron to result in formation of ion radicals or ROS such as superoxide (O₂˙⁻) and hydroxyl radicals (˙OH). In type II process, ³PS reacts directly with ³O₂ through transfer of energy to result in formation of ¹O₂.³ It is generally accepted that the ¹O₂ is the major species responsible for PDT efficacy of most PSs.³ Although, PSs which acts via type I process have also been considered useful for PDT.⁴

Currently, PSs such as Photofrin, ALA, Foscan and Verteporfin have been clinically approved and some more PSs are under clinical trials.¹ For improving, the tumor selectivity and efficacy of PDT, PSs complexed to metal ions have been actively explored.⁵ The insertion of metal in PS provides two significant advantages. First, the metal through heavy atom effect can enhance triplet state yield of the PS and second because of cationic charge the PS can accumulate more efficiently in cancer cells.^5,6 In addition, metal based PS have received considerable attention for potential use as multimodal agents in tumor imaging and therapy such as magnetic resonance imaging (MRI), positron emission tomography (PET) imaging,^7,8 X-ray photon activation therapy (PAT),^9–11 and radiotherapy of cancer.¹²

Our earlier studies in hamster cheek pouch model have shown that Cp₆, a water soluble chlorophyll derivative is a promising PS for PDT of tumors.^13,14 Cp₆ is anionic due to the presence of carboxylic groups in the molecule. For further developing its use as a multimodal agent for PDT and PAT, we have explored the possibility to obtain its metal complex by chelation of metal ion at the carboxylic groups. We choose CuI to prepare complex with Cp₆ because of several considerations; copper complexes are potential agents for chemo and radiotherapy of cancer,^15,16 iodinated compounds are useful for PAT and contrast imaging,^17,18 the radioisotopes of both copper and iodine can provide modality for PET and single-photon emission computed tomography (SPECT) imaging of tumor,^19,20 and both copper and iodine are considered to be relatively well tolerated metabolically because these are required as trace elements.²¹ In this report, we describe the synthesis of an iodinated–Chlorin p₆ copper complex (ICp₆–Cu) and its characterization as a PS. Unlike the tetrapyrrole copper complex reported in literatures which are metallated at the pyrrole NH, the copper in ICp₆–Cu is attached to the carboxylic groups. Studies on ¹O₂ and ROS generation showed that ICp₆–Cu predominantly acts through type I photochemical pathway and as compared to Cp₆, it is more effective for PDT of cancer cells under hypoxic condition.

Results and discussion

Formation of complex of Cp₆ with CuI

The changes in the absorption spectrum of Cp₆ upon addition of different concentrations of CuI are shown in Fig. 1a. With the increase in the concentration of CuI, the absorbance of Q band at 663 nm gradually decreases with concomitant appearance of a new band at 634 nm. This is also accompanied by a decrease in absorbance at the Soret band (400 nm) and the minor Q band (500 nm). Beyond 5 μM concentration of CuI which is equi-molar to Cp₆, no further changes were observed. This indicates that the complex formation between Cp₆ and CuI is completed at 1 : 1 stoichiometry. Complex thus formed displays a Soret band with sharp peak at 408 nm with a shoulder at 380 nm, a minor Q_x band at 500 nm and the major Q_y band at 634 nm. The molar extinction coefficient (ε) of the complex at Q_y band is almost same (24 000 M⁻¹ cm⁻¹) as that for Cp₆, whereas at the Soret band the ε is ∼30% lower (63 000 M⁻¹ cm⁻¹). In Fig. 1b the fluorescence spectra of Cp₆ and ICp₆–Cu complex are shown. As compared to Cp₆, the fluorescence yield for ICp₆–Cu is reduced by a factor of ∼16. This is because of insertion of paramagnetic copper in Cp₆ and is consistent with the quenching of fluorescence reported for copper complex of some chlorophyll derivatives.²² X-Ray fluorescence (XRF) spectrum of the complex (Fig. 2) shows the characteristic X-ray emission lines for copper at 8.03 keV (K_α) and 8.9 keV (K_β1) and for iodine at 3.97, 4.25, 4.5, 4.81 keV (L_α1, L_β1, L_β2, L_γ1). This also confirms the presence of both copper and iodine in the complex.

Fig. 1 (a) Changes in absorption spectra of 5 μM Cp₆ solid line (black) with subsequent increase in concentration of CuI in methanol; 1 μM (red), 2 μM (blue), 3 μM (purple), 4 μM (pink), 5 μM (orange), 10 μM (green). (b) Fluorescence spectra (λ_ext = 400 nm) of Cp₆ (solid line) and ICp₆–Cu (dashed line) in methanol.

Fig. 2 XRF spectrum of ICp₆–Cu showing characteristic X-ray emission lines for copper and iodine (inset). X-ray emission lines for copper at 8.03 keV (K_α) and 8.9 keV (K_β1) and for iodine at 3.97, 4.25, 4.5, 4.81 keV (L_α1, L_β1, L_β2, L_γ1). The peak at 2.97 keV and 3.2 keV is due to atmospheric argon and the peak at 1.75 keV is due to silicon wafer on which powdered sample is deposited.

Position of copper and iodine in the complex

Generally, metallation of tetrapyrrole compounds leads to insertion of metal atom in the centre of the tetrapyrrole ring by substitution with two hydrogen at the pyrrole NH. However for tetrapyrrolic compound containing carboxylic groups, it has been reported that the insertion of metal ion at pyrrole NH is not favoured due to preferential binding of metal ion at the carboxylic groups through coordination.²³ Cp₆ contains three carboxylic groups, one is as propionate side chain at C17 position and remaining two are adjacently attached at C13 and C15 position.

To confirm the coordination site of copper in ICp₆–Cu, its Electron Paramagnetic Resonance (EPR) spectrum was measured at 100 K in methanol (Fig. 3). The spectrum is anisotropic with characteristic intense absorption in high field region and well resolved hyperfine splitting in low field region. It is important to mention that the EPR spectra of copper complexes of porphyrin and chlorophyll derivatives show characteristic multiple super hyperfine splitting in g_⊥ region due to interaction of unpaired electron of metal with neighbouring nitrogen nucleus.²⁴ Such hyperfine splitting is absent in the EPR spectrum of ICp₆–Cu which suggests that copper is not coordinated to pyrrole nitrogen. This is in agreement with a previous report on platinum(III) complex of hematoporphyrin where the EPR spectra of the complex having platinum attached to tetrapyrrole ring and the one in which platinum is attached to side chain carboxylic groups show similar difference.²⁵ Further, the value of g-tensor and hyperfine splitting can be used to infer the coordination of copper with oxygen, nitrogen or sulphur based on the Peisach–Blumberg correlation.²⁶ The value of g_∥ (2.27) and A_∥ (174 × 10⁻⁴ cm⁻¹) obtained from the EPR spectrum are within the range of copper coordinating with oxygen atom. These findings unequivocally confirm that copper is attached to carboxylic groups. The nature of metal ligand bond can be predicted by calculating covalency parameter α² from the following equation.

α_Cu² = −(A_∥/0.036) + (g_∥ − 2.003) + 3/7(g_∥ − 2.003) + 0.04 (1)

Fig. 3 EPR spectrum of ICp₆–Cu at 100 K in methanol.

A value of α² = 0.5 indicates complete covalent nature and if the value is 1.0 it indicates ionic bonding.²⁷ The value of α² for ICp₆–Cu is 0.82 which indicates that the nature of bonding between ligand and copper has some covalent character. Moreover, the value of α² is higher than that reported for copper chlorophyll derivatives for Cu–N bond (0.63–0.68). This also suggests that the nature of copper–ligand bond in ICp₆–Cu is significantly different.²⁴

In Fig. 4, we show the FTIR spectra of ICp₆–Cu and Cp₆. The FTIR spectrum of Cp₆ is very similar to the FTIR spectrum of Chlorin e6 reported by Gladkova et al.²⁸ This is to be expected because the chemical structure of Ce6 is similar to Cp₆ except the presence of one extra methylene (–CH₂–) group at C15 position. The FTIR spectra of Cp₆ in the region from 1300 cm⁻¹ to 1750 cm⁻¹ shows absorption bands due to C=O and COO stretching vibrations of the carboxylic groups (Fig. 4b). The major changes observed upon complex formation are as follows (1) appearance of three new bands for ICp₆–Cu, one at 1637 cm⁻¹ which is characteristic of CO–metal stretch^29,30 and two prominent bands at 1112 cm⁻¹ and 1203 cm⁻¹ due to CO stretching. In a previous report, the appearance of later two bands have been shown as direct proof for the formation of chelate of copper and zinc with C13 keto group of chlorophyll.³⁰ (2) The COO vibration bands in Cp₆ arise as a prominent band resolved into a peak at 1558 cm⁻¹ and shoulders at 1578 cm⁻¹ and 1596 cm⁻¹. This band in ICp₆–Cu is broad and one of the band at 1578 cm⁻¹ disappeared. (3) The CO vibrations of C13 and C15 carboxylic group (at 1419 cm⁻¹ and 1321 cm⁻¹) are shifted to lower frequency at 1396 cm⁻¹ and 1313 cm⁻¹, respectively. These observations suggest that the two adjacent carboxylic groups at C13 and C15 position participate in formation of the metal coordination complex. The third carboxylic group at C17 propionic side chain is likely to remain attached to sodium because the region ∼1740 cm⁻¹ which is characteristic of C=O vibrational band of this carboxylate show no change. In addition to above noted changes, the FTIR spectrum of ICp₆–Cu also revealed an intense broad absorption band at ∼3400 cm⁻¹ (Fig. 4c) which is expected to appear due to the presence of water molecules in the co-ordination sphere of copper.²⁹

Fig. 4 FTIR spectra of Cp₆ (black) and ICp₆–Cu (red) plotted in the range of (a) 600–1250 cm⁻¹, (b) 1300–1750 cm⁻¹ and (c) 2800–4000 cm⁻¹.

The mass spectrum of ICp₆–Cu is shown in Fig. 5. The molecular ion containing copper is identified by the characteristic isotopic splitting at 724.12 m/z and 726.12 m/z which represent loss of C17 carboxylate from the molecule. The largest peak observed at 701 m/z shows no such spliting and represents molecular ion with loss of metal and vinyl side group. This also rules out the possibility of iodination at vinyl group of Cp₆. Previous reports on iodination of chlorophyll derivatives have shown that of the three meso positions at C5, C10, and C20, the meso position at C20 is relatively more reactive towards halogenation.³¹ The selective iodination of chlorophyll-a derivatives at C20 position has been reported in several studies.^31–33 The replacement of meso-H with halogen in tetrapyrrole ring is identified by disappearance of characteristic proton peak in the ¹H NMR spectrum. However, because the presence of paramagnetic copper leads to broadening of lines, a well resolved ¹H NMR spectrum of ICp₆–Cu could not be obtained. Based on evidence gathered from XRF (Fig. 2), EPR (Fig. 3), FTIR (Fig. 4), and HRMS (Fig. 5), the proposed chemical structure of ICp₆–Cu is shown in Fig. 6. The site of iodination is indicated as most probable based on literature discussed above and needs to be confirmed.

Fig. 5 Mass spectrum of ICp₆–Cu in the range 550–800 m/z, showing molecular ion at 701 m/z (after loss of Cu and a vinyl (C₂H₃) from parent molecule) and spectrum magnified in the range 720–750 (inset) to show peaks of isotopic ion at 724.12 m/z and 726.12 m/z due to presence of natural isotopes of copper ⁶³Cu and ⁶⁵Cu (after loss of –COONa from parent molecule).

Fig. 6 Proposed structure of ICp₆–Cu. The probable site of iodine is indicated by dashed bond.

¹O₂ and ROS generation efficiency of ICp₆–Cu

The ¹O₂ and ROS generation efficiency of ICp₆–Cu was measured using three different fluorescence probes Singlet Oxygen Sensor green (SOSG), 2′,7′-dichlorodihydrofluorescin (DCFH) and 3′-(p-aminophenyl) fluroscein (APF). SOSG is specific for ¹O₂,³⁴ DCFH detects total ROS³⁵ and APF can detect ˙OH as well as ¹O₂.³⁶ The ¹O₂ or ROS generated in photochemical reaction converts these probes into highly fluorescent species. The change in fluorescence intensity (F_t/F₀) of SOSG and DCF as a function of irradiation time were plotted to compare the ¹O₂ and ROS generation capability of Cp₆ (Fig. 7a) and ICp₆–Cu (Fig. 7b). For ICp₆–Cu, the relative increase in SOSG fluorescence is ∼10 times less than that for Cp₆ (Fig. 7a and b), which suggest that as compared to Cp₆, the efficiency of ICp₆–Cu to produce ¹O₂ is reduced. It may be noted here that while the presence of heavy atom in the PS can improve the yield of ¹O₂ by increasing the efficiency of ISC from ¹PS* to the ³PS,^5,37 any shortening of excited state life time would result in a loss of ¹O₂ generation ability of the PS.³⁸ Copper being paramagnetic results in reduction in excited state lifetime because of which most of the copper complex of porphyrin and chlorophyll derivatives reported in literature are photodynamically inactive.^22,39 In contrast to copper, the presence of iodine on the PS usually increase the yield of ³PS and results in enhancement in the ¹O₂ yield.^40,41 However, iodine depending on its position on the PS molecule can also results in shortening of ³PS life time and decrease in ¹O₂ yield.³⁷ This is due to the fact that the orbital overlap between tetrapyrrole ring and iodine atom facilities non-radiative decay of the ³PS.³⁷ Thus the decrease in ¹O₂ yield of ICp₆–Cu can be attributed to the presence of both copper and iodine. It is also pertinent to note that PS with short life time of ³PS can be photodynamically active via type I process. For example, copper octaethylbenzochlorin despite short ³PS life time induce efficient photosensitization via type I process and has been found useful for PDT of cancer both in in vitro and in vivo studies.⁴²

Fig. 7 Cp₆ (a) and ICp₆–Cu (b) induced photodynamic generation of ¹O₂ and total ROS plotted as increase in fluorescence (F_t/F₀) of SOSG (upside arrow) and DCF (circle), respectively. The solution of PS and fluorescence probe in phosphate buffer (pH 7.4) was irradiated with red light (3.5 mW cm⁻²) for different time periods. Each data point represents mean ± SD value of three independent experiments.

The total ROS generation ability of ICp₆–Cu was measured using fluorescence probe dichlorofluorescein (DCF). Fig. 7b show changes in DCF fluorescence induced by photoactivation of Cp₆ or ICp₆–Cu as a function of irradiation time. Here, the increase in DCF fluorescence is higher in Cp₆ than for ICp₆–Cu. However, since the yield of ¹O₂ for ICp₆–Cu is low, the increase in DCF fluorescence is mainly due to generation of other ROS. Whereas for Cp₆, the increase in DCF fluorescence is contributed predominantly by generation of ¹O₂ as seen in SOSG assay (Fig. 7b). These results show that the generation of other ROS by ICp₆–Cu is substantially higher than the generation of ¹O₂. To confirm this, increase in DCF fluorescence was monitored in the presence of Sodium azide (NaN₃), a quencher of ¹O₂. However, it was observed that the addition of NaN₃ to DCF solution led to alteration in fluorescence of DCF which interfered with the measurement. In a previous report, fluorescence probe APF together with NaN₃ and D₂O (enhances life time of ¹O₂) has been effectively used to distinguish type I or type II photochemical mechanism of PDT drug.³⁶ For Cp₆ and ICp₆–Cu, the changes in fluorescence of APF (F_t/F₀) as a function of irradiation time is shown in Fig. 8a and b. For Cp₆, the increase in fluorescence of APF is inhibited to ∼50% and ∼90% in the presence of 5 mM and 10 mM NaN₃, respectively and it is enhanced by a factor of ∼4 in D₂O (Fig. 8a). These results show that ¹O₂ is predominant species in photodynamic action of Cp₆ which is also consistent with our previous report.⁴³ In contrast, for ICp₆–Cu induced increase in APF fluorescence is not significantly influenced by either NaN₃ or D₂O (Fig. 8b). These results clearly show that the photodynamic action of ICp₆–Cu is mediated predominantly via type I process.

Fig. 8 Cp₆ (a) and ICp₆–Cu (b) induced photodynamic generation of ROS plotted as increase in fluorescence (F_t/F₀) of APF in H₂O (square), D₂O (circle) and in H₂O containing NaN₃ at 5 mM (upward arrow) and 10 mM (downward arrow) concentration. The solution of PS and fluorescence probe in phosphate buffer (pH 7.4) was irradiated with red light (3.5 mW cm⁻²) for different time periods. Each data point represents mean ± SD value of three independent experiments.

Phototoxicity of ICp₆–Cu in cancer cells

The dependence of phototoxicity of ICp₆–Cu on the concentration and the light dose in two oral cancer cell lines (4451 and NT8e) is shown in Fig. 9A and B, respectively. Treatment of cells with various concentration of ICp₆–Cu followed by light exposure at fixed light dose of 7 J cm⁻² led to concentration dependent increase in phototoxicity (Fig. 9A). At 10 μM ICp₆–Cu, the phototoxicity was ∼80% which increased further upto ∼95% at higher concentrations. However, there was also some (∼10%) dark toxicity in both cell lines beyond 10 μM ICp₆–Cu (see ESI Table S1†). Based on this, 10 μM concentration of ICp₆–Cu was selected to study the dependence of phototoxicity on light dose. As shown in Fig. 9B, PDT with 10 μM ICp₆–Cu led to a dose dependent increase in phototoxicity. The phototoxicity was ∼60% at light dose of 5 J cm⁻² and increased to ∼90% at 12 J cm⁻². The changes in cell viability observed after staining with LIVE/DEAD kit is shown in Fig. 9C. In both NT8e and 4451, while all cells in control appear green (Fig. 9C(a) and(c)), the cells after PDT take up the red fluorescence due to loss of cell viability (Fig. 9C(b) and (d)). The morphological changes in two cell lines after PDT are shown in Fig. 10A. The cells in control appear healthy with intact cell membrane (Fig. 10A(a) and (c)). At 3 h after PDT (LD₉₀ dose), almost all NT8e cells show formation of membrane blebs whereas, 4451 cells display damaged plasma membrane and swelling, indicated by arrows (Fig. 10A(b) and (d)). The percentages of live, apoptotic and necrotic cells in control and PDT treated cells for the two cell lines is shown in Fig. 10B. As expected, the mode of cell death in NT8e and 4451 cells occurred mainly via apoptosis and necrosis, respectively. This is due to the fact that the tumor suppressor p53 gene responsible for apoptotic death is wild type in NT8e cells whereas it is mutated in 4451 cells.⁴⁴ These results are consistent with our earlier report on the mode of cell death induced by Cp₆ and its histamine conjugate in these cell lines.⁴⁴ The influence of quenchers of ¹O₂ (NaN₃, L-histidine) and free radicals (dimethyl sulfoxide (DMSO), D-mannitol) on ICp₆–Cu induced phototoxicity is shown in Fig. 11. As compared to control, the level of phototoxicity is significantly reduced in the presence of all the quenchers. Treatment of cells with quenchers alone under similar conditions did not result in any toxicity. It is important to mention that NaN₃ and histidine can also quench ˙OH radicals.⁴⁵ However, since the rate constant of NaN₃ and histidine for reaction with ˙OH radicals is almost same, both should lead to inhibition of phototoxicity to the same extent. In contrast, the inhibition of phototoxicity by NaN₃ is higher than that for histidine (Fig. 11) which correlates with their efficiency to quench ¹O₂.⁴⁵ This suggests that in addition to free radicals, ¹O₂ also contributes to the phototoxicity of ICp₆–Cu. The contribution of ¹O₂ to the phototoxicity of ICp₆–Cu was not expected because studies on ROS generation in aqueous solution showed that the ¹O₂ generation ability of ICp₆–Cu is very low (Fig. 7b and 8b). Here it is pertinent to note that the efficacy of the PS to generate ¹O₂ also depends on its environment. In aqueous solution, the PSs with higher hydrophobicity tend to aggregate and show decrease in ¹O₂ generation efficiency whereas the localization of PS in cellular membrane results in dis-aggregation and increase in photosensitization via type II process.^46,47 A comparison of the absorption spectra of ICp₆–Cu in aqueous medium and methanol (see ESI Fig. S1†) suggested that it is slightly aggregated in aqueous condition as indicated by significant blue shift and broadening of soret and Q bands (see ESI Table S2†). Further, the octanol : water partition coefficient (Log P) of ICp₆–Cu is 0.94 which is ∼2 times higher than Cp₆ (see ESI Table S3†). Therefore due to higher hydrophobicity, ICp₆–Cu can localize efficiently in cellular membrane.⁴⁸ This and the fact that the life time of ¹O₂ in cellular environment is considerably longer (10 μs) as compared to that in aqueous media (3.5 μs)⁴⁹ can explain the contribution of ¹O₂ to the phototoxicity for ICp₆–Cu.

Fig. 9 (A) Concentration dependent phototoxicity of ICp₆–Cu in NT8e (open circle) and 4451 (open downward arrow) at fixed light dose of 7 J cm⁻². (B) Phototoxicity of ICp₆–Cu in NT8e (open circle) and 4451 (open downward arrow) cell lines as a function of light dose, (C) fluorescence microphotograph of NT8e and 4451 cells in control (a, c) and at 24 h after PDT (b, d) with light dose of 8 J cm⁻² showing live and dead cells after staining with live/dead fluorescence probe, scale bar – 20 μm.

Fig. 10 (A) Microphotograph of NT8e and 4451 cells in control (a, c) and at 3 h after PDT (b, d) showing cell morphology in phase contrast, magnification 20×, scale bar – 20 μm. Arrows showing formation of apoptotic blebs in NT8e (b), and plasma membrane damage and swelling in 4451 (d). (B) Percentage of live, apoptotic and necrotic cells at 3 h after PDT in NT8e (white bar) and 4451 (shaded bar) cell lines. Light dose used in A and B, is 8 J cm⁻². Data are mean ± SD of three independent experiments in triplicates.

Fig. 11 Percent phototoxicity induced by ICp₆–Cu in NT8e cells in absence and presence of quenchers of ROS. The cells were subjected to PDT using light dose of 4 J cm⁻² which led to ∼50% phototoxicity in absence of any quencher. Data are mean ± SD of three independent experiments in triplicates.

Phototoxicity under hypoxic conditions

Availability of molecular oxygen is an important factor for the PDT efficacy of majority of PSs. It has been reported that low oxygen concentration can reduce the ¹O₂ yield and compromise the PDT efficacy.^50–52 Alternatively, studies on PDT of human tumor xenografts with meta-tetrahydroxyphenylchlorin have shown that photochemical process leading to radical formation can work better under hypoxic condition.⁵² Several recent studies also showed that PS which act via type I photochemical mechanism induce better PDT efficacy under hypoxia. For example, phototoxicity mediated by methylene blue encapsulated in polymer nanoparticle⁵³ and 5,10,15,20-tetrakis(meso-hydroxyphenyl)porphyrin incorporated in electron-rich poly(2-(diisopropylamino)ethyl methacrylate) micelles is not affected under hypoxia.⁵⁴ Similarly, photodynamic release of doxorubicin induced by chondroitin sulfate conjugated Pheophorbide-a via type I mechanism led to higher toxicity under hypoxic conditions than that under normoxic conditions.⁵⁵

The effect of hypoxia on the phototoxicity of ICp₆–Cu and Cp₆ at LD₅₀ and LD₈₀ dose (for normoxic condition) is shown in Fig. 12. For Cp₆, hypoxia led to decrease in phototoxicity by almost 1/2 (∼30% and 40%). In contrast, phototoxicity induced by ICp₆–Cu under normoxic and hypoxic condition show no significant difference. Thus, as compared to Cp₆, ICp₆–Cu led to higher phototoxicity under hypoxic condition. It is important to mention that hypoxia has also been shown to affect the phototoxicity of bacteriopheophorbide which acts only via oxygen radicals, whereas the phototoxicity of Type II PS Npe₆ was found to be unaffected.⁵⁶ The possible reason for PDT efficacy of Npe₆ under hypoxia was referred to superior ability of its ³PS to interact with oxygen.⁵⁶ Although in our studies, as compared to Cp₆ which also acts mainly via ¹O₂, the ability of ICp₆–Cu to act via type I process appear important for its PDT efficacy under hypoxic condition.

Fig. 12 Percent phototoxicity of ICp₆–Cu and Cp₆ in NT8e cells after PDT under normoxia (white bar) and hypoxia (shaded bar). The light dose used corresponds to LD₅₀ and LD₈₀ for cells under normoxic condition. Data are mean ± SD of three independent experiments in triplicates.

Conclusions

To summarize, we have synthesized iodinated Chlorin p₆ copper complex for PDT applications. EPR and FTIR studies suggest that copper is attached to the two adjacent carboxylic groups through co-ordination and the site of iodination may be meso position in the tetrapyrrole ring. Results of studies on the photodynamic efficacy of ICp₆–Cu show that it is able to induce high phototoxicity in cancer cells and the phototoxicity is not affected significantly under hypoxic condition. In conclusion, results suggest that the novel copper complex is useful for PDT of cancer.

Experimental section

Preparation of Cp₆ and ICp₆–Cu

Chlorophyll-a was extracted from dried spirulina powder and converted into Cp₆ following the procedure described by Hoober et al.⁵⁷ Cp₆ was converted into ICp₆–Cu according to our patent specifications (no. 4912/MUM/2015). In brief, the solution of Cp₆ and CuI in methanol : acetonitrile (1 : 10 v/v) was mixed at equimolar concentration at room temperature. The solvent was partially evaporated to result in precipitation of the metal complex. The precipitates were collected by centrifugation (6000g, 10 min). The metal complex was further purified by silica column chromatography using methanol : acetonitrile solvent system. The stock solution of ICp₆–Cu was prepared in ethanol : PEG (400) : water (20 : 30 : 50) to prevent aggregation and contamination of the concentrated solution during storage.

Spectroscopy

The UV/VIS absorption spectra of the Cp₆ and ICp₆–Cu were recorded from 350–750 nm, at 1 nm bandpass on a Cintra-20 spectrophotometer (GBC, Australia). The fluorescence emission spectra of Cp₆ and ICp₆–Cu were recorded on a Fluorolog-2 spectroflurometer (Spex, USA) by setting excitation wavelength at ∼400 nm, scan range from 500–700 nm and band pass of ∼1.7 nm and ∼3.7 nm for excitation and emission slits, respectively. FTIR spectra were recorded on a spectrometer model (FTLA 2000 MB 104) using powdered sample deposited on Zinc Selenide window. The X band EPR spectrum of ICp₆–Cu was recorded on Spectrometer model JES – (Jeol, Japan) at 100 K. Methanol was used as solvent.

The X-ray fluorescence of ICp₆–Cu was monitored using microprobe XRF beamline (BL-16) of the synchrotron radiation source (Indus-2).⁵⁸ For preparation of sample for XRF, 10 μl concentrated solution of ICp₆–Cu in methanol was applied on a piece of Silicon wafer and air dried. Monochromatic X-ray of energy 12.0 keV (beam size 4 mm × 4 mm) was used to excite the fluorescence from the sample. XRF spectra were recorded using a Vortex energy-dispersive detector (SII NanoTechnology, USA).

Cell culture

Human oral cancer cell lines NT8e derived from upper aerodigestive tract (pyriform fossa) and 4451 derived from recurrent tumor in the lower jaw were obtained from ACTREC, Mumbai and INMAS, Delhi respectively. Both the cell lines were maintained in Dulbecco's Modified Eagle's Medium (DMEM) containing 4.5 g L⁻¹ glucose, 4 mM L-glutamine, 25 mM of HEPES, antibiotics (streptomycin, nystatin, penicillin) and 10% fetal bovine serum. The cells were grown at 37 °C in a humidified incubator (Galaxy R, RS Biotech) under 5% CO₂ + 95% air atmosphere. The cells were harvested by trypsinization and seeded in 35 mm Petri dishes or 96 well plate. After 18 h of incubation, the cells in exponential phase were used for the experiments.

Irradiation

For assay of ROS generation and phototoxicity with ICp₆–Cu, the irradiation was done using an LED lamp (Dia 4 cm) emitting red light (630 ± 20 nm). To ensure homogenous illumination, a diffuser glass plate was placed below the LED lamp. For similar studies with Cp₆, a diode laser source (Thorlabs, USA) emitting red light (660 ± 5 nm) was used. Power of light was measured using a power meter model AN/2 (Ophir) and the power density was kept nearly same (∼3.5 mW cm⁻²) for both the light source.

Generation of ¹O₂ and ROS

To determine the ¹O₂ and other ROS yield for ICp₆–Cu and Cp₆ we used three different fluorescence probes; SOSG and APF both from Invitrogen (USA) and 2′,7′-dichlorodihydrofluorescin-diacetate (DCFH-DA) from Sigma, USA. SOSG reacts specifically with ¹O₂ to result in formation of highly fluorescent endoperoxide product.³⁴ DCFH-DA is a fluorescence probe for detection of intracellular generation of broad range of ROS such as O₂˙⁻ and ˙OH, hydrogen peroxide (H₂O₂) and ¹O₂. In cells, DCFH-DA is first converted enzymatically into DCFH which upon reaction with ROS is oxidized into green fluorescent DCF. For detection of ROS in solution, DCFH-DA is chemically converted into DCFH by hydrolysis according to the protocol described by Bourré et al.³⁵ For this, DCFH-DA was mixed with 0.01 N NaOH in absolute ethanol and the solution was kept at room temperature for 30 min to allow hydrolysis. After neutralization with 25 mM sodium phosphate buffer (pH 7.4), the solution was stored on ice in dark until use. The fluorescence probe APF has been used to detect ˙OH and ¹O₂ both.³⁶ APF upon reaction with ˙OH and ¹O₂ is converted to a fluorescent form due to the cleavage of the aminophenyl ring from fluorescein ring system.

For monitoring generation of ¹O₂ and other ROS, ICp₆–Cu or Cp₆ (5 μM) was added to 3.5 ml sodium phosphate buffer (pH 7.4) in a plastic petridish (3.5 cm dia). After addition of the fluorescence probe, the solution was irradiated with narrow bandwidth light of appropriate wavelengths 630 ± 20 nm or 660 ± 5 nm for photoactivation of ICp₆–Cu or Cp₆ respectively, using light source described in section ‘Irradiation’. The irradiation was done in diffused dim light and the solution was continuously stirred during irradiation. At different time interval, 200 μl aliquot was withdrawn and fluorescence was measured on a spectrofluorometer. The wavelength used for fluorescence excitation of SOSG, DCF and APF were 485 nm, 488 nm and 490 nm respectively. The change in fluorescence intensity at 530 nm, 525 nm and 515 nm for SOSG, DCF and APF respectively were plotted as F_t/F₀ vs. irradiation time, where F₀ and F_t are the fluorescence intensity before and after irradiation.

Photosensitizer treatment

The growth medium from 35 mm petri dishes or 96 well plate containing cell monolayer was removed and replaced with fresh media containing ICp₆–Cu. After incubation for 3 h at 37 °C in a humidified CO₂ incubator in dark, the growth medium was aspirated and the cell monolayer was washed twice using plain medium (DMEM without serum). After adding fresh growth medium, the cells were subjected to photodynamic treatment. For concentration dependent PDT experiments, concentration of ICp₆–Cu was varied from 2–15 μM and cells were irradiated with a fixed light dose of 7 J cm⁻². For experiments where the light dose was varied, cells were treated with fixed concentration of 10 μM and irradiated with different light doses (0–12.7 J cm⁻²). For PDT under hypoxia, a fresh growth medium which was de-oxygenated by bubbling nitrogen gas for 1 h was used. The de-oxygenated medium was added to the cells after PS treatment and the cells were transferred to humidified incubator (ESCO) under controlled environment (1% O₂ + 5% CO₂) at 37 °C for 1 h. To maintain the hypoxia during irradiation, the 96 well plate containing cells was placed in a sealed glass chamber and the chamber was flushed with nitrogen. For studies with quenchers of ROS (NaN₃, histidine, DMSO and mannitol), the cells prior to the photodynamic treatment were incubated for 1 h in growth medium containing specified concentration of each quencher and then irradiated at LD₅₀ light dose.

MTT assay

The dark and light induced cytotoxicity was determined using MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay.⁵⁹ Briefly, at 24 h after PDT, the growth medium of the treated cells was replaced with plain medium containing 0.5 mg ml⁻¹ MTT and the cells were incubated at 37 °C for 3 h in dark. After incubation, the medium was removed and 100 μl of DMSO was added and kept for 15 min to solubilise the formazan crystals formed within the cells. A microplate reader (Power Wave 340, Bio-tek instruments Inc., USA) was used to read the absorbance of the samples at 570 nm with reference wavelength at 690 nm. The percent phototoxicity was calculated with respect to the absorbance value in control samples. The cells treated with ICp₆–Cu but not exposed to light were used as dark control. In each experiment at least four replicates were used and mean value was used to determine the percent cytotoxicity.

Cell morphology and viability

PDT-induced cell damage and loss of cell viability was examined by bright field and fluorescence microscopy at 20× magnification using inverted microscope (Olympus, Japan). For cell viability, the cells after 24 h of PDT were stained with LIVE/DEAD cytotoxicity kit (Invitrogen, Molecular Probes, Eugene OR). The kit contains a green fluorescent dye calcein, which is retained in the cytoplasm by live cells and a DNA binding red fluorescent dye ethidium homodimer, that enter the cells only upon loss of plasma membrane integrity. After staining the cells were washed with phosphate buffered saline (PBS) and observed under fluorescence microscope using blue (450–480 nm) and green excitation (510–550 nm).

Assessment of apoptosis and necrosis

To determine the percentage of cells undergoing apoptosis and necrosis, the cells after PDT were stained with Hoechst 33 342 (10 μg ml⁻¹) and propidium iodide (PI, 2.5 μg ml⁻¹) for 5 min, washed twice with PBS and observed under inverted microscope (Olympus, Japan). PI is a cell impairment dye and can only stain the cells in which plasma membrane integrity is lost due to necrotic damage. Hoechst 33 342 is cell permeable, which stains nuclei and allows visualization of chromatin condensation and fragmentation of apoptotic cells. The blue fluorescence of Hoechst 33342 and red fluorescence of PI was visualized using excitation with 380–400 nm, barrier filter 440 nm. Images were recorded using a color CCD camera model ‘Prog Res Cfscan’ and a ProgRes Capture Pro Software (Jenoptik, Germany). A minimum of 500 cells were counted in each group and percentage of apoptotic, necrotic and live cells was calculated from the total number of cells counted.

Disclosure

Alok Dube, Paromita Sarbadhikary and Pradeep Kumar Gupta are named patent inventors for Indian Patent Application no. 4912/MUM/2015 titled ‘A metal complex of chlorophyll derivative for magnetic resonance imaging and photodynamic therapy’, filed on December 29, 2015.

Acknowledgements

We are thankful to Dr G. S. Lodha and Dr M. K. Tiwari for helping with XRF measurement, Dr B. Jain for helping with FTIR measurements at our centre and Dr M. Shanmugam, IIT Bombay, Mumbai for EPR measurement. We also thank Dr B. S. Dwarakanath, INMAS, Delhi and Dr U. M. Warawdekar, ACTREC, Mumbai for providing oral cancer cell lines. S Paromita thanks Homi Bhabha National Institute, Mumbai for senior research fellowship.

References

1. P. Agostinis, K. Berg, K. A. Cengel, T. H. Foster, A. W. Girotti, S. O. Gollnick, S. M. Hahn, M. R. Hamblin, A. Juzeniene, D. Kessel, M. Korbelik, J. Moan, P. Mroz, D. Nowis, J. Piette, B. C. Wilson and J. Golab, Ca-Cancer J. Clin., 2011, 61, 250–281.
2. R. R. Allison and K. Moghissi, Photodiagn. Photodyn. Ther., 2013, 10, 331–341.
3. A. P. Castano, T. N. Demidova and M. R. Hamblin, Photodiagn. Photodyn. Ther., 2004, 1, 279–293.
4. Y. M. Riyad, S. Naumov, S. Schastak, J. Griebel, A. Kahnt, T. Häupl, J. Neuhaus, B. Abel and R. Hermann, J. Phys. Chem. B, 2014, 118, 11646–11658.
5. L. B. Josefsen and R. W. Boyle, Met.-Based Drugs, 2008, 2008, 276109.
6. L. Benov, Med. Princ. Pract., 2015, 24, 14–28.
7. R. K. Pandey, N. S. James, Y. Chen, J. Missert and M. Sajjad, Methods Mol. Biol., 2010, 635, 223–259.
8. H. U. Rashid, M. N. Umar, K. Khan, M. N. Anjum and M. Yaseen, J. Struct. Chem., 2014, 55, 910–915.
9. J. Gastaldo, Z. Bencokova, C. Massart, A. Joubert, J. Balosso, A. M. Charveta and N. Foraya, J. Synchrotron Radiat., 2011, 18, 456–463.
10. T. Ishizumi, K. Aizawa, T. Tsuchidaa, T. Okunakaa and H. Katoa, Photodiagn. Photodyn. Ther., 2004, 1, 295–301.
11. M. Miura, G. M. Morris, J. W. Hopewell, P. L. Micca, M. S. Makar, M. M. Nawrocky and M. W. Renner, Br. J. Radiol., 2012, 85, 443–450.
12. P. A. Waghorn, J. Labelled Compd. Radiopharm., 2014, 57, 304–309.
13. A. Dube, S. Sharma and P. K. Gupta, Oral Oncol., 2006, 42, 77–82.
14. A. Dube, S. Sharma and P. K. Gupta, Oral Oncol., 2011, 47, 467–471.
15. C. Santini, M. Pellei, V. Gandin, M. Porchia, F. Tisato and C. Marzano, Chem. Rev., 2014, 114, 815–862.
16. G. Hao, A. N. Singh, O. K. Oz and X. Sun, Curr. Radiopharm., 2011, 4, 109–121.
17. R. W. Miller, W. DeGraff, T. J. Kinsella and J. B. Mitchell, Int. J. Radiat. Oncol., Biol., Phys., 1987, 13, 1193–1197.
18. J. M. Widmark, Proc. Bayl. Univ. Med. Cent., 2007, 20, 408–417.
19. C. J. Anderson and R. Ferdani, Cancer Biother. Radiopharm., 2009, 24, 379–393.
20. B. Gutfilen and G. Valentini, BioMed Res. Int., 2014, 2014, 426892.
21. L. Prashanth, K. K. Kattapagari, R. T. Chitturi, V. R. Baddam and L. K. Prasad, J. NTR Univ. Health Sci., 2015, 4, 75–85.
22. L. M. Moreira, A. Lima, R. R. S. Soares, V. R. Batistela, A. P. Gerola, N. Hioka, J. A. Bonacin, D. Severino, M. S. Baptista, A. E. H. Machado, M. R. Rodrigues, L. Codognotoa and H. P. M. Oliveira, J. Braz. Chem. Soc., 2009, 20, 1653–1658.
23. E. B. Fleischer, E. I. Choi, P. Hambright and A. Stose, Inorg. Chem., 1964, 3, 1284–1287.
24. Y. Nonomura, N. Yoshioka and H. Inoue, Inorg. Chim. Acta, 1994, 224, 181–184.
25. G. Momekov, M. Karaivanova, I. Ugrinova, E. Pasheva, G. Gencheva, D. Tsekova, S. Arpadjan and P. R. Bontchev, Invest. New Drugs, 2011, 29, 742–751.
26. S. K. Hoffmann, J. Goslar, S. Lijewski and A. Zalewska, J. Magn. Reson., 2013, 236, 7–14.
27. N. Raman, S. Ravichandran and C. Thangaraja, J. Chem. Sci., 2004, 116, 215–219.
28. O. L. Gladkova, M. V. Parkhats, A. N. Gorbachova and S. N. Terekhov, Spectrochim. Acta, Part A, 2010, 76, 388–394.
29. M. Q. Zhang, Y. C. Zhu, J. G. Wu, P. Shi, R. W. Deng and Z. N. Chen, Chem. Pap., 2001, 55, 202–205.
30. J. Petrovic, G. Nikolic and D. Markovic, J. Serb. Chem. Soc., 2006, 71, 501–512.
31. M. Ethirajan, P. Joshi, W. H. William, K. Ohkubo, K. S. Fukuzumi and R. K. Pandey, Org. Lett., 2011, 13, 1956–1959.
32. H. Tamiaki, N. Arik, S. Yasuda, T. Miyatak and T. Oba, Tetrahedron, 2014, 70, 9768–9775.
33. H. Tamiaki, Y. Kotegawa and K. Mizutani, Bioorg. Med. Chem. Lett., 2008, 18, 6037–6040.
34. A. Gollmer, J. Arnberg, F. H. Blaikie, B. W. Pedersen, T. Breitenbach, K. Daasbjerg, M. Glasius and P. R. Ogilby, Photochem. Photobiol., 2011, 87, 671–679.
35. L. Bourré, S. Thibaut, A. Briffaud, N. Rousset, S. Eléouet, Y. Lajat and T. Patrice, J. Photochem. Photobiol., B, 2002, 67, 23–31.
36. M. Price, J. J. Reiners, A. M. Santiago and D. Kessel, Photochem. Photobiol., 2009, 85, 1177–1181.
37. F. Nifiatis, J. C. Athas, K. D. D. Gunaratne, Y. Gurung, K. M. Monette and P. J. Shivokevich, Open Spectrosc. J., 2011, 5, 1–12.
38. S. Mathai, T. A. Smith and K. P. Ghiggino, Photochem. Photobiol. Sci., 2007, 6, 995–1002.
39. M. C. DeRosa and R. J. Crutchley, Coord. Chem. Rev., 2002, 233, 351–371.
40. A. C. Serra, M. Pineiro, A. M. Rocha Gonsalves, M. Abrantes, M. Laranjo, A. C. Santos and M. F. Botelho, J. Photochem. Photobiol., B, 2008, 92, 59–65.
41. S. H. Lim, C. Thivierge, P. Nowak-Sliwinska, J. Han, H. van den Bergh, G. Wagnières, K. Burgess and H. B. Lee, J. Med. Chem., 2010, 53, 2865–2874.
42. G. M. Garbo, V. H. Fingar, T. J. Wieman, E. B. Noakes 3rd, P. S. Haydon, P. B. Cerrito, D. H. Kessel and A. R. Morgan, Photochem. Photobiol., 1998, 68, 561–568.
43. B. Bose and A. Dube, J. Photochem. Photobiol., B, 2008, 93, 32–35.
44. A. Parihar, A. Dube and P. K. Gupta, Cancer Chemother. Pharmacol., 2011, 68, 359–369.
45. S. Basu-Modak and R. M. Tyrrell, Cancer Res., 1993, 53, 4505–4510.
46. L. I. Grossweiner, A. S. Patel and J. B. Grossweiner, Photochem. Photobiol., 1982, 36, 159–167.
47. H. Mojzisova, S. Bonneau, P. Maillard, K. Berg and D. Brault, Photochem. Photobiol. Sci., 2009, 8, 778–787.
48. F. M. Engelmann, S. V. Rocha, H. E. Toma, K. Araki and M. S. Baptista, Int. J. Pharm., 2007, 329, 12–18.
49. J. Baier, M. Maier, R. Engl, M. Landthaler and W. Bäumler, J. Phys. Chem. B, 2005, 109, 3041–3046.
50. J. Moan and S. Sommer, Cancer Res., 1985, 45, 1608–1610.
51. J. D. Chapman, C. C. Stobbe, M. R. Arnfield, R. Santus, J. Lee and M. S. McPhee, Radiat. Res., 1991, 126, 73–79.
52. V. O. Melnikova, L. N. Bezdetnaya, D. Brault, A. Y. Potapenko and F. Guillemin, Int. J. Cancer, 2000, 88, 798–803.
53. M. Usacheva, S. K. Swaminathan, A. R. Kirtane and J. Panyam, Mol. Pharm., 2014, 11, 3186–3195.
54. H. Ding, H. Yu, Y. Dong, R. Tian, G. Huang, D. A. Boothman, B. D. Sumer and J. Gao, J. Controlled Release, 2011, 156, 276–280.
55. W. Park, B. C. Bae and K. Na, Biomaterials, 2016, 77, 227–234.
56. M. Price, L. Heilbrun and D. Kessel, Photochem. Photobiol., 2013, 89, 683–686.
57. J. K. Hoober, T. W. Sery and N. Yamamoto, Photochem. Photobiol., 1988, 48, 579–582.
58. M. K. Tiwari, P. Gupta, A. K. Sinha, S. R. Kane, A. K. Singh, S. R. Garg, C. K. Garg, G. S. Lodha and S. K. Deb, J. Synchrotron Radiat., 2013, 20, 386–389.
59. T. Mosmann, J. Immunol. Methods, 1983, 65, 55–63.

---

Footnote

† Electronic supplementary information (ESI) available: Table S1: percent cytotoxicity of ICp₆–Cu in dark at 10–15 μM concentration; Fig. S1: absorption spectra of ICp₆–Cu in buffer and methanol; Table S2: Absorption characteristics of ICp₆–Cu in methanol and buffer; Table S3: octanol/water partition coefficient of Cp₆ and ICp₆–Cu. See DOI: 10.1039/c6ra14026b

---