Photodegradation of Nonoxynol-9 in Aqueous Media.
Studies of its in vitro phototoxicity
N. Canudas* and C. Rivas
Laboratorio de Fotoquímica, Centro de Química, Instituto Venezolano de Investigaciones Científicas, Apartado 21827, Caracas 1020-A, Venezuela.
Received 27 June 1997
Copyright © 1997 Internet Journal of Science - Biological Chemistry
The photodegradation of Nonoxynol-9 was studied under aerobic conditions. It involves cleavage of the ethereal chain.
Nonoxynol-9 is phototoxic in vitro as indicated by the photohemolysis test. Furthermore Nonoxynol-9 photosensitises peroxidation of linoleic acid as monitored by the UV detection of dienic hydroperoxides. Partial inhibition of these processes on addition of butylated hydroxyanisole (BHA) and reduced glutathione (GSH) suggests the involvement of type I mechanism.
Nonoxynols are used as non-ionic surface active agents, acting as surfactants in numerous cosmetic products and antiseptic preparations. Furthermore, Nonoxynol -9 and -10 are also commonly used in spermicidal contraceptive products that have been recommended in the prevention of sexually transmitted diseases and in human immunodeficiency virus profilaxis. They have been found to be irritants or sensitisers and their photosensitising effect has been widely recognised [1-3]. During the administration of these agents allergic contact dermatitis and photosensitising reactions have been observed [2,4]. This motivated us to study the photochemistry of Nonoxynol-9 as a model of this type of surfactants. The aims of this paper were: a) the photodegradation of nonoxynol-9 (1) in aqueous medium under aerobic conditions trying to identify the major photoprocesses; b) the in vitro phototoxicity of Nonoxynol-9 using human erythrocytes (photohemolysis) and photodynamic lipid peroxidation on linoleic acid; and finally c) to postulate a mechanism for the nonoxynol phototoxicity.
Figure 1. Chemical structure of Nonoxynol-9
2. Investigations and results
The absorption spectrum of nonoxynol-9 in buffered aqueous medium shows two bands centred at 225 and 276 nm with a tail extending to 300 nm. These spectral characteristics indicate that the drug has the prerequisites to act as UVB sensitizer. The compound 1 was found to be photolabile in PBS towards UVB light (290-320 nm) and in aerobic medium. Monitoring the spectral changes of the absorption bands of 1 (1 x 10-4 M) at regular intervals of 5 minutes irradiation of aerated solutions (Fig. 2) little spectral variations occurred in the first period (below 35 min), suggesting that photoproducts with an absorption spectrum very similar to that of the starting 1 were formed. At irradiation times longer than 35 min the spectrum changed with a new maximun growing to 300 nm.
Figure 2. UV monitoring of the photolysis of Nonoxynol-9 (10-4 M) in PBS under aerated
conditions at regular intervals of 5 min of irradiation.
After 48 h the phototransformation of 1 (1 x 10-3 M) was completed as observed by HPLC chromatogram (Fig. 3). Photolysis of 1 at preparative scale and purification by thin layer chromatography allowed to separate two types of photoproduct mixtures. Under deaerated conditions the degradation process was very slow to be detected in an hour.
Figure 3. A: Chromatogram of Nonoxynol-9; B: Chromatogram of photoproducts
Red blood cells (RBCs) suspensions containing 1 (2 x 10-5 M) undergo photohemolysis when irradiated under aerobic conditions (Fig. 4). No lysis was observed when cells were irradiated for 80 min in the absence of 1 as well as when RBCs were incubated with 1 in the dark.
The effects of the additives were inferred by considering the percentage of photohemolysis attained in the same amount of time used to reach 60% photohemolysis in the absence of additives (Fig. 4). It was found that GSH and BHA can inhibit the process in about 95%, whereas sodium azide and DABCO showed only 50% and 30% protective effect respectively in the process. Comparative photohemolysis in the presence of additives after 60 min of irradiation is shown in figure 5.
Figure 4. Photohemolysis of RBCs sensitised by 1 (2 x 10-5 M) under oxygen, argon and
in the presence of additives (5 x10-5M).
Figure 5. Comparative photohemolysis in the presence of Nonoxynol-9 and additives after
60 min of irradiation.
Due to the damaging effects of photoperoxidation to cell membranes, this process is thought to play an important role in skin phototoxicity. When a PBS solution of linoleic acid (10-3 M) was irradiated in the presence of 1 (10-5M), significant amounts of dienic hydroperoxides were formed, as evidenced by the appearance of a new UV-absorption band at 233 nm (Fig. 6).
Figure 6. Uv-vis of the linoleic acid (10-3 M in PBS) photoperoxidation; I: 3h irradiation;
II: Linoleic acid +1, 1h irradiation; III: Linoleic acid +1, 3h irradiation.
The increase of the UV absorption band (l=233 nm) was spectrophotometrically followed at 1 and 3 hours time, since it allowed a quantitative stimation of the final hydroperoxide concentration (between 10-4 and 10-5 M) on the basis of the known extinction coefficient (e = ca. 31000).
Homolytic rupture of the ethereal chain is the probable mechanism of photodegradation of 1 since the spectroscopic data showed a pattern corresponding to a mixture of compounds with a similar structure of 1 for the fraction with lower Rf while for the fraction with higher Rf only an aliphatic ethereal structure was observed by NMR. The general mechanism of photochemical reactions for aromatic ethers  is a rearrangement to obtain a phenolic compound. Probably, Nonoxynol-9 undergoes similar rupture to obtain a complex mixture of photoproducts, as observed by HPLC and 1HNMR, scheme 1. A radical mechanism is in accord with the hemolysis results obtained for 1 in the presence of RBCs.
The fact that the photodegradation rate of 1 under deaerated conditions was slow is in agreement with the fact that the photohemolysis was significantly reduced in argon atmosphere. The presence of oxygen is apparently important for photodegradation to take place in spite that the presence of sodium azide decreases in 50% the photohemolysis observed respect to that under oxygenated conditions. GSH and BHA can inhibit in 95% the photohemolysis and it is in agreement with the mechanism of radicals proposed by which the radicals formed can attack the membrane of the erythrocytes.
The lipid photoperoxidation shows an important correlation with the damage produced in cell membranes and therefore with the resulting skin phototoxicity . The observed photohemolysis by 1 might be reflecting extensive photoperoxidation of the membrane lipids. At present it has been found in this laboratory that 1 photosensitises the peroxidation of linoleic acid (Fig. 6). A possible mechanism of lipid photoperoxidation based on a radical chain process (type I mechanism) is confirmed by the well correlated inhibition of the photohemolytic processes by BHA and GSH, well stabilised radical scavengers .
In summary, Nonoxynol-9 is photolabile in PBS under aerobic conditions. The radicals formed during its photodegradation can cause lysis of the erythrocyte membrane and can induce peroxidation of linoleic acid. Perhaps with an increasing use of nonoxynols as components of antiseptic lotions  many more photoreactions will be seen in the future. A careful use of these compounds should be recommended to patients.
Scheme 1. Photodegradation of Nonoxynol-9 in PBS under aerobic conditions
Nonoxynol-9 (1) was extracted from Norforms®, (NorPharm Laboratory C.A., Venezuela) by dissolving in warm ethanol and cooling the solution in an ice bath; petrolatum was precipitated and the ethanol solution containing the nonoxynol was separated. The purity of Nonoxynol-9 was 99% as determined by 1H NMR-spectroscopy (Bruker Aspect 3000, 300 MHz). Reduced glutathione (GSH), butylated hydroxyanisole (BHA) and linoleic acid were purchased from Sigma (Saint Louis, MO, USA). Sodium azide (NaN3) and 1,4-diazabicyclo [2.2.2] octane (DABCO) were purchased from Aldrich.
Nonoxynol-9 (1) was irradiated at room temperature for 48 h in PBS (Phosphate Buffered Saline Solution, pH 7.4) (1 x 10-4M), using a quartz immersion well photoreactor (Applied Photophysics parts no. 3230+3307) and a Rayonet photochemical chamber reactor model RPR-100 equipped with 16 phosphor RPR-lamps with an emission maximum at 300 nm, under aerated conditions. The course of the reactions was followed by UV-Vis spectrophotometry using a Milton Roy 3000 instrument, by monitoring the changes in its absorption bands at regular intervals of 5 minutes of irradiation, as shown for a PBS solution (10-4 M) of 1 in Fig. 2. The course of the reaction was also followed by HPLC (Waters Delta Prep 4000 equipped with a Hypersil ODS semipreparative column 100 x 4,6 mm2, 5 mm packing), using methanol-water gradient as mobil phase at a flow rate of 1,2 ml min-1, with monitoring at 222 nm. All solvents were analytical or spectra-grade.
When irradiation was completed, the PBS solution was extracted with CHCl3 (3 x 100 ml); the organic phase was evaporated at reduced pressure and the residue was purified by thin layer chromatography (TLC) using precoated silica gel F254 plates (Merck) and methanol/chloroform, (80:20) as eluent. Two types of photoproduct mixtures were isolated which were analysed by 1H NMR spectroscopy. The fraction with lower Rf, Rf=0,5 (60%) showed the following spectroscopic data: 1H MNR (CDCl3, 300 MHz): d = 6.9-7.4 (m, aromatic-H), 1.0-4.0 (m, aliphatic H) and the fraction with higher Rf, Rf=0,8 (40%) showed: 1H MNR (CDCl3, 300 MHz) d= 2.5-4.0 (m, aliphatic-H).
A suspension of red blood cells (RBCs) from freshly obtained human erythrocytes was prepared by washing them four times with a tenfold volume of a phosphate-buffered saline solution (PBS, pH 7.4, 0.01 M phosphate buffer and 0.135 M NaCl), centrifuging the cells each time at 2500 g for 15 min and carefully removing the supernatant. For the hemolysis experiments, RBCs were diluted in PBS containing compound 1 until the resulting suspension had an optical density (OD) of 0.4-0.8 at 650 nm. An OD value of 0.5 corresponded to 3.3 x 106 cells ml-1. This was read on a Milton-Roy 3000 spectrophotometer. The hemolysis rate and hemolysis percentage were determined by measuring the decreasing OD at 650 nm, since the OD is proportional to the number of intact RBCs [8,9]. The sample containing the drug in concentration of 2 x 10-5 M was irradiated under oxygen and argon enriched atmosphere in a Rayonet Photochemical Reactor equipped with 16 phosphorous lamps with an emission maximum at 300 nm for a period of 70 min. Anaerobic conditions were obtained by bubbling argon through the PBS solution before adding the erythrocytes. A similar experiment was performed with 1 in the presence of radical traps (GSH, BHA) and singlet oxygen quenchers (NaN3 and DABCO)  in concentrations of 5 x 10-5 M. All the data are the average (mean arithmetic) of the values obtained by repeating the experiments three times. Control experiments were performed to ruled out possible interferences due to hemolysis occurring in the dark.
4.4 Photosensitised Peroxidation of Linoleic Acid:
Linoleic acid (10-3 M in PBS ), was irradiated in the presence of the compound 1 (10-5 M), and the formation of dienic hydroperoxides was monitored by UV-spectrophotometry, through the appearance and progressive increase of a new band at l = 233 nm [6,11,12] . The measurements were made at time of 1 and 3 hours of irradiation.
1 Dooms-Goossens, A.; Deveylder, H.; Gidi de Alam, A.; Lachapelle, J.
M.; Tennstedt, D.; Degreef, H.: J. Am. Acad. Dermatol. 21, 723 (1989)
2 Christian, M. S.; ed. Cosmetic ingredient review: final report on the
safety assessment of nonoxynols -2, -4, -8, -9, -10, -12, -14, -15, -30, -40
and -50 : J. Am..Coll. Toxicol. 2, 35 (1983) .
3 Nethercott J. R.; Lawrence M. J.: Contact Dermatitis 10, 235 (1984).
4 Michel, M.; Dompmartin, A.; Moreau, A.; Leroy, D.: Photodermatol.
Photoimmunol. Photomed. 10, 198 (1994).
5 Kelly, D. P.; Pinhey, J. T.; Rigby, R. D. G.: Australian Journal of
Chemistry 22, 977 (1969).
6 Giroti, A. W.: Photochem. Photobiol. 51, 497 (1990).
7 Moore, D. E.; Chapuis, P. P.: Photochem. Photobiol. 47, 173 (1988).
8 Valenzeno, D. P.; Trank, J. W.: Photochem. Photobiol. 42, 335 (1985).
9 Miranda, M. A.; in: Castell, I. V. (Ed.): Alternative in vitro models to
animal experimentation in pharmaco toxicology; p. 239, Farmaindustria,
10 Costanzo, L. L.; De Guidi, G.; Condorelli, G.; Cambria, A.; Fama, M.:
Photochem. Photobiol. 50, 359 (1989).
11 Recknagal, R. O.; Glende, E. A. Jr.; in: Packer, L. (Ed.): Oxygen
Radicals in Biological Systems, Methods in Enzymology; V. 105,
p.331, Academic Press, New York 1984.
12 Giuffrida, S.; De Guidi, G.; Sortino, S.; Chillemi, R.; Costanzo, L. L.;
Condorelli, G.: J. Photochem. Photobiol. B: Biology 29,125 (1995).
This research was supported by grants from the CONICIT ( SI-2502 ) and Fundación Central Madeirense C.A. (Caracas, Venezuela).
Address correspondence to: Nieves Canudas Crespo, Laboratorio de Fotoquímica, Centro de Química, Instituto Venezolano de Investigaciones Científicas, Apartado 21827, Caracas 1020-A, Venezuela. Fax number: 58(2)5041350, E-mail: firstname.lastname@example.org