Phthalocyanine-induced photohemolysis: structure-activity relationship and the effect of fluoride

Singlet Oxygen Quantum Yields of Different Photosensitizers in Polar Solvents and Micellar Solutions

The singlet oxygen luminescence method and the photochemical methods using 1,3-diphenylisobenzofuran ( DPBF ) or bilirubin ditaurate ( BDT ) as chemical quenchers were employed to determine the single oxygen quantum yields (ΦΔ) of different phthalocyanines and tris(2,2″-bipyridyl)ruthenium(II) dichloride in dimethylformamide ( DMF ) or aqueous micellar solution of 0.1 M CTAC (cetyltrimethylammonium chloride). Additionally, a perylenetetracarboxylic acid diimide derivative was examined in DMF . In a series of tetrasulfonated phthalocyanines ( PTS ) the following order was found: ZnPTS > GaPTS > AlPTS ≈ H 2 PTS > CoPTS . In general, the singlet oxygen quantum yields are higher in DMF than in 0.1 M CTAC/H 2 O . The results obtained with the photochemical systems are comparable with those obtained by the photophysical method.

The photochemical DPBF method results in absolute values of ΦΔ. However, in micellar solution, chain reactions occur when DPBF is used as chemical quencher in the photo-oxidative process. This problem can be overcome by adding sodium thiosulphate which is able to destroy the endoperoxide initially formed. BDT can be used as quencher in different solvents to determine ΦΔ relative to a photosensitizer with known singlet oxygen quantum yield in the respective solvent.

In comparison to the chemical methods the luminescence method exhibits the advantage that side reactions of the quencher are excluded. But normally the ΦΔ values obtained are relative to a reference, since absolute determinations need much larger efforts.

Singlet Oxygen Quantum Yields of Different Photosensitizers in Polar Solvents and Micellar Solutions

The singlet oxygen luminescence method and the photochemical methods using 1,3-diphenylisobenzofuran ( DPBF ) or bilirubin ditaurate ( BDT ) as chemical quenchers were employed to determine the single oxygen quantum yields (ΦΔ) of different phthalocyanines and tris(2,2″-bipyridyl)ruthenium(II) dichloride in dimethylformamide ( DMF ) or aqueous micellar solution of 0.1 M CTAC (cetyltrimethylammonium chloride). Additionally, a perylenetetracarboxylic acid diimide derivative was examined in DMF . In a series of tetrasulfonated phthalocyanines ( PTS ) the following order was found: ZnPTS > GaPTS > AlPTS ≈ H 2 PTS > CoPTS . In general, the singlet oxygen quantum yields are higher in DMF than in 0.1 M CTAC/H 2 O . The results obtained with the photochemical systems are comparable with those obtained by the photophysical method.

The photochemical DPBF method results in absolute values of ΦΔ. However, in micellar solution, chain reactions occur when DPBF is used as chemical quencher in the photo-oxidative process. This problem can be overcome by adding sodium thiosulphate which is able to destroy the endoperoxide initially formed. BDT can be used as quencher in different solvents to determine ΦΔ relative to a photosensitizer with known singlet oxygen quantum yield in the respective solvent.

In comparison to the chemical methods the luminescence method exhibits the advantage that side reactions of the quencher are excluded. But normally the ΦΔ values obtained are relative to a reference, since absolute determinations need much larger efforts.

Light-triggered liposomal release: membrane permeabilization by photodynamic action

Photosensitized damage to liposome membranes was studied by using different dye-leakage assays based on fluorescence dequenching of a series of dyes upon their release from liposomes. Irradiation of liposomes with red light in the presence of a photosensitizer, trisulfonated aluminum phthalocyanine (AlPcS(3)), resulted in the pronounced leakage of carboxyfluorescein, but rather weak leakage of sulforhodamine B and almost negligible leakage of calcein from the corresponding dye-loaded liposomes. The same series of selectivity of liposome leakage was obtained with chlorin e6 that appeared to be more potent than AlPcS(3) in bringing about the photosensitized liposome leakage. Electrically neutral zinc phthalocyanine tetrasubstituted with a glycerol moiety (ZnPcGlyc(4)) was less effective than negatively charged AlPcS(3) in provoking the light-induced liposome permeabilization. On the contrary, both ZnPcGlyc(4) and AlPcS(3) were much more effective than chlorin e6 in sensitizing gramicidin channel inactivation in planar bilayer lipid membranes, thus showing that relative photodynamic efficacy of sensitizers can differ substantially for damaging different membrane targets. The photosensitized liposome permeabilization was apparently associated with oxidation of lipid double bonds by singlet oxygen as evidenced by the mandatory presence of unsaturated lipids in the membrane composition for the photosensitized liposome leakage to occur and the sensitivity of the latter to sodium azide. The fluorescence correlation spectroscopy measurements revealed marked permeability of photodynamically induced pores in liposome membranes for such photosensitizer as AlPcS(3).

Photodynamic activity and binding of sulfonated metallophthalocyanines to phospholipid membranes: Contribution of metal-phosphate coordination

Photosensitized efficacy of tetrasulfonated phthalocyanines of zinc, aluminum and nickel (ZnPcS(4), AlPcS(4) and NiPcS(4), respectively) as studied by gramicidin channel (gA) photoinactivation was compared with adsorption of the dyes on the surface of a bilayer lipid membrane as measured by the inner field compensation method. The adsorption of the negatively charged phthalocyanines on diphytanoylphosphatidylcholine (DPhPC) membranes led to formation of a negative boundary potential difference between the membrane/water interfaces. Good correlation was shown between the photodynamic activity and the membrane binding of the three metallophthalocyanines. ZnPcS(4) appeared to be the most potent of these photosensitizers, while NiPcS(4) was completely ineffective. All of these phthalocyanines displayed no binding and negligible gA photoinactivation with membranes formed of glycerol monooleate (GMO), whereas Rose Bengal exhibited significant binding and photodynamic efficacy with GMO membranes. Gramicidin photoinactivation in the presence of AlPcS(4), being insensitive to the ionic strength of the bathing solution, was inhibited by fluoride and attenuated by phosphate ions. A blue shift of the fluorescence peak position of ZnPcS(4) dissolved in ethanol was elicited by phosphate, similarly to fluoride, which was indicative of the coordination interaction of these ions with the central metal atom of the phthalocyanine macrocycle. This interaction was enhanced in the medium modeling the water-membrane interface. The results obtained imply that binding of tetrasulfonated metallophthalocyanines to phospholipid membranes is determined primarily by metal-phosphate coordination.

The mechanism of Zn-phthalocyanine photosensitized lysis of human erythrocytes

The phthalocyanines have recently been suggested as one of most effective possible sensitizers for photodynamic therapy and the blood viral inactivation. The further characterisation of the mechanism of human red blood cell lysis and membrane alterations upon photodynamic treatment in the presence of Zn-phthalocyanine was the aim of this study. It was found that there were (2.7+/-0.4).10(7) dye binding sites per red blood cell with the association constant equal to (1.4+/-0.3).10(4) M(-1). Two types of the photosensitized haemolysis: haemolysis during irradiation ("light" haemolysis) and post-irradiation haemolysis ("dark" haemolysis) were studied. The erythrocyte membrane hyperpolarisation, membrane fluidisation and cell swelling preceded the "light" haemolysis. The modification of the erythrocyte membrane band 3 protein by DIDS (an inhibitor of anion exchange) increased the rate of the "light" haemolysis. The rate of "dark" haemolysis was higher and that of "light" haemolysis was lower in potassium media in comparison to sodium ones. The rates of photohaemolysis depended on the erythrocyte membrane potential: a decrease of membrane potential inhibited both types of haemolysis. The cell shrinkage in the presence of sucrose (up to 15 mM) inhibited the "dark" haemolysis but significantly increased the "light" haemolysis. Oxidation of intracellular oxyHb to metHb by nitrite, which drastically decreases intracellular oxygen concentration, as well as GSH concentration, inhibited the rate of the "light" haemolysis. The results allow for the conclusion that the mechanism of photochemical ("light") haemolysis is not of a colloid-osmotical type, in contrast to the post-irradiation ("dark") haemolysis. The photochemical oxidation or denaturation of band 3 protein plays a significant role in the formation of haemolytic holes. The membrane lipid peroxidation, as well as glutathione oxidation, does not participate in the process of photosensitized haemolysis. From the inhibition of "dark" haemolysis by sucrose the apparent pore radius was estimated to be about 1.1 nm. The pores appear to be transient short-lived ones, the average pore number per cell was 0.02.

Effect of fluoride anions on gramicidin photoinactivation sensitized by sulfonated aluminum phthalocyanines

Interaction of potent photodynamic agents, sulfonated aluminum phthalocyanines (AlPcSn where n is a number of sulfonic groups), with biological membranes was studied here using model systems: sensitized photoinactivation of gramicidin channels in planar lipid bilayers and adsorption on lipid monolayers. Fluoride anions known to form complexes with aluminum were found to inhibit both the adsorption of aluminum phthalocyanines on lipid monolayers, as measured with a Langmuir trough by surface pressure and surface potential changes, and photodynamic efficacy of the dyes, as studied by gramicidin channel photoinactivation. The similar effects were caused by the alkalinization of the medium. Fluoride anions appeared to be much more effective in the case of AlPcS4 as compared to AlPcS3. The suppression of the photodynamic potency of aluminum phthalocyanines was attributed to desorption of the dyes from lipid bilayers induced by fluoride or hydroxyl ions. With AlPcS4 an enhancement of the dye aggregation leading to a decrease in the sensitizing activity was probably involved in the fluoride effect as revealed by absorption and fluorescence spectral measurements. Capillary electrophoresis was employed to understand the mechanism of the dye desorption. The results of these experiments indicated that the reduction in the membrane affinity was associated with an increase in the negative charge of the dye molecules due to the binding of fluoride or hydroxyl ions.

Photosensitizer binding to lipid bilayers as a precondition for the photoinactivation of membrane channels

The photodynamic activity of sulfonated aluminum phthalocyanines (AlPcS(n), 1 </= n </= 4) was found to correlate with their affinity for membrane lipids. Adsorbing to the surface of large unilamellar vesicles (LUVs), aluminum phthalocyanine disulfonate induced the highest changes in their electrophoretic mobility. AlPcS(2) was also most efficient in mediating photoinactivation of gramicidin channels, as revealed by measurements of the electric current across planar lipid bilayers. The increase in the degree of sulfonation of phthalocyanine progressively reduced its affinity for the lipid bilayer as well as its potency of sensitizing gramicidin channel photoinactivation. The portion of photoinactivated gramicidin channels, alpha, increased with rising photosensitizer concentration up to some optimum. The concentration at which alpha was at half-maximum amounted to 80 nM, 30 nM, 200 nM, and 2 microM for AlPcS(1), AlPcS(2), AlPcS(3), and AlPcS(4), respectively. At high concentrations alpha was found to decrease, which was attributed to quenching of reactive oxygen species and self-quenching of the photosensitizer triplet state by its ground state. Fluoride anions were observed to inhibit both AlPcS(n) (2 </= n </= 4) binding to LUVs and sensitized photoinactivation of gramicidin channels. It is concluded that photosensitizer binding to membrane lipids is a prerequisite for the photodynamic inactivation of gramicidin channels.

The influence of merocyanine 540 and protoporphyrin on physicochemical properties of the erythrocyte membrane

The interaction of the red cell membrane with merocyanine 540 or protoporphyrin led to four phenomena, most probably interrelated. (i) The morphology changed from the normal discoid to an echinocytic form. This morphological change persisted when followed over a period of 24 h. (ii) Simultaneously, cell deformability was decreased, as revealed by viscosity measurements and a cell-filtration technique. (iii) Both drugs caused swelling of the erythrocytes in isotonic medium, due to a very-short-term increased permeability of the membrane, also for larger molecules such as lactose. The pathway of this temporary leak seems to be unrelated to the Na+/K+ -ATPase, the K+/Cl- and the Na+/K+/Cl- cotransport systems, the Ca2+-activated Gardos pathway, the oxidation/deformation-activated leak pathway and the so-called residual transport route. Despite the morphological changes, K+-leakage induced by mechanical stress was not increased. (iv) During osmotic swelling, the critical hemolytic volume was found to be increased in the presence of either merocyanine 540 or protoporphyrin. The increase critical volume protected erythrocytes against osmotic hemolysis.

Elevation of GRP-78 and loss of HSP-70 following photodynamic treatment of V79 cells: sensitization by nigericin

Chinese hamster V79 cells were treated with photodynamic therapy (PDT) sensitized by aluminum phthalocyanine (AlPc) or with the ionophore nigericin or with combinations of PDT and nigericin. We previously showed that PDT and nigericin interact synergistically in the killing of these cells; i.e. doses of PDT that kill no more than 10% of the cells in combination with nontoxic exposures to nigericin lead to a loss of clonogenicity of three to five orders of magnitude. Photodynamic therapy induces an enhanced rate of expression of the stress gene grp-78 both at the transcriptional and translational levels and causes a decrease in the synthesis of the constitutive heat shock protein HSP-70 as well as in expression of HSP-70 mRNA. When the cells are exposed to PDT in the presence of nigericin, these effects are elicited at three- to four-fold lower PDT doses. Thus, PDT in the presence of nigericin is much more effective in inducing the changes in gene expression than is PDT alone. In the absence of nigericin the PDT dose inducing a two-fold increase in GRP-78 accumulation causes little or no loss of clonogenicity. In the presence of nigericin, however, the PDT dose leading to a similar change in GRP-78 level produces up to a 50% loss of clonogenicity. The fact that nigericin is dose-modifying for both cell killing and stress responses suggests that nigericin either increases the yield of oxidative damage from a given dose of PDT or magnifies the cellular response to a constant level of oxidative stress.

The effect of irradiance on virus sterilization and photodynamic damage in red blood cells sensitized by phthalocyanines

Phthalocyanines are being studied as photosensitizers for virus sterilization of red blood cells (RBC). During optimization of the reaction conditions, we observed a marked effect of the irradiance on production of RBC damage. Using a broad-band light source (600-700 nm) between 5 and 80 mW/cm2, there was an inverse relationship between irradiance and rate of photohemolysis. This effect was observed with aluminum sulfonated phthalocyanine (AlPcSn) and cationic silicon (HOSiPc-OSi[CH3]2[CH2]3N+[CH3]3I- phthalocyanine (Pc5) photosensitizers. The same effect occurred when the reduction of RBC negative surface charges was used as an endpoint. Under the same treatment conditions, vesicular stomatitis virus inactivation rate was unaffected by changes in the irradiance. Reduction in oxygen availability for the photochemical reaction at high irradiance could explain the effect. However, theoretical estimates suggest that oxygen depletion is minimal under our conditions. In addition, because the rate of photohemolysis at 80 mW/cm2 was not increased when irradiations were carried out under an oxygen atmosphere this seems unlikely. Likewise, formation of singlet oxygen dimoles at high irradiances does not appear to be involved because the effect was unchanged when light exposure was in D2O. While there is no ready explanation for this irradiance effect, it could be used to increase the safety margin of RBC virucidal treatment by employing exposure at high irradiance, thus minimizing the damage to RBC.

The photochemical properties of fluoroaluminum phthalocyanine

Fluoride is known to inhibit the photodynamic activity of aluminium phthalocyanine in a variety of biological systems. In order to gain insight into this phenomenon, the effect of fluoride on the photophysical properties of free and albumin-bound chloroaluminum phthalocyanine sulfonate (AlPcSn) were studied. The association constant of NaF with AlPcSn in aqueous solution was measured as 500 +/- 20 M-1. This binding affects the photophysical properties of the dye: the absorption bands in the visible range are blue-shifted by 6-8 nm, and this effect is mirrored in the fluorescence emission spectrum. Human serum albumin significantly quenched the dye fluorescence independent of the presence of fluoride ion. The transient absorption spectrum of the excited dye triplet is unchanged by NaF, but the quantum yield for its generation is increased by 50%, with no decrease in its lifetime. Formation of fluoroaluminum phthalocyanine complexes was also observed in tetrabutylammonium fluoride-assisted solutions in wet acetonitrile. The fluoro-AlPcSn complex is a better photosensitizer for generation of singlet oxygen than the original dye-hydroxyl ion complex, as confirmed using the imidazole-N,N-dimethyl-4-nitrosoaniline method. On the other hand, the fluoro-AlPcSn complex exhibits an intense inhibitory effect on photohemolysis of red blood cells (RBC) even after the cells are washed to remove free dye and fluoride prior to irradiation, indicating that once the dye is attached to the cellular site, the fluoride ligand is no longer prone to displacement (by hydroxyl ion, for example). Nonetheless, it is clear from the spectroscopic data that the new fluoro complex is an efficient sensitizer for photooxidation.(ABSTRACT TRUNCATED AT 250 WORDS)

New phthalocyanines for photodynamic virus inactivation in red blood cell concentrates

Cationic phthalocyanines with either aluminum or silicon as the central metal were evaluated for their ability to inactivate viruses in red blood cell concentrates (RBCC) photodynamically. In addition, the virucidal potential of a substituted anionic phthalocyanine, aluminum dibenzodisulfophthalocyanine hydroxide (A1N2SB2POH) was evaluated and compared with that of the much studied anionic aluminum tetrasulfophthalocyanine hydroxide (A1PcS4OH). Based on the rate of inactivation of the lipid-enveloped vesicular stomatitis virus (VSV), the virucidal potential of these phthalocyanines was: HOSiPcOSi(CH3)2(CH2)3N+(CH3)3I- (Pc 5) = SiPc[OSi(CH3)2-(CH2)3N+(CH3)3I-]2 (Pc 6) > A1PcOSi(CH3)2(CH2)3N+(CH3)2(CH2)11CH3I- (Pc 21) = A1N2SB2POH = A1PcS4 > HOSiPc[OSi(CH3)2(CH2)3N+(CH3)2(CH2)11CH3I-]2 (Pc 14) > A1PcOSi(CH3)2(CH2)3N+(CH3)3I- (Pc 2). Phthalocyanine ligand 14 and Pc 21 are new phthalocyanines, made by quaternizing known amino analogues. Compared to VSV, the rate of inactivation of Sindbis virus (another model lipid-enveloped virus) was identical when treated in red blood cells (RBC) with Pc 5 and slightly higher when treated with Pc 6 and A1PcS4OH. Treatment of RBCC containing cell-free human immunodeficiency virus (HIV-1) with Pc 5 or A1PcS4OH required 15 min of irradiation to inactivate (> 5 log10 reduction) the virus. The extent of HIV-1 inactivation with A1N2SB2POH was 3.7 log10 after 60 min of red light exposure. The RBC integrity after photosensitization was measured by the ability of the cells to bind to plates coated with poly-L-lysine, (which reflects the retention of the RBC surface negative charges) and hemolysis of the cells over a 7 day storage period.(ABSTRACT TRUNCATED AT 250 WORDS)

Photodynamic inactivation of herpes viruses with phthalocyanine derivatives

The antiviral photosensitization capacity of 11 different phthalocyanine (Pc) derivatives was examined using herpes simplex virus-1, herpes simplex virus-2 and varicella zoster virus in the search for the most potent sensitizers for viral decontamination of blood. The kinetics of viral photoinactivation were resolved during the stages of viral adsorption and penetration into the host cells. The capacity of Pc in the photodynamic inactivation of viruses was compared with that of merocyanine 540 (MC540), another widely studied photosensitizer. Sensitivity to photoinactivation decreased progressively with time after addition of viruses to their host cells. The viruses were most sensitive to photodynamic inactivation up to 30 min from the initiation of adsorption. Cell-associated viruses, 45-60 min after the onset of adsorption, are highly resistant to photodynamic treatment by most photosensitizers, with the exception of amphiphilic Pc derivatives. Thus the mixed sulfonated Pc-naphthalocyanine derivatives AlNSB3P and AlN2SB2P demonstrated a remarkable decontamination activity even 60 min after the onset of adsorption. Ultrastructural examination of these photosensitized viruses demonstrated damage to the viral envelope which prevented viral adsorption and/or penetration. The non-enveloped adenovirus was found to be resistant to all the dyes tested.