Abstract
Aims
To study reaction of photoactivated frusemide (F) and F glucuronide (Fgnd metabolite) with human serum albumin in order to find a clue to clarify a mechanism of phototoxic blisters from high frusemide dosage.
Methods
F was exposed to light in the presence of human serum albumin (HSA). HSA treated with this method (TR-HSA) was characterized by fluorescence spectroscopic experiment, alkali treatment and reversible binding experiment.
Results
Less 4-hydroxyl-N-furfuryl-5-sulphamoylanthranilic acid (4HFSA, a photodegradation product of F) was formed in the presence of HSA than in the absence of HSA. A new fluorescence spectrum excited at 320 nm was observed for TR-HSA. Alkali treatment of TR-HSA released 4HFSA. Quenching of the fluorescence due to the lone tryptophan near the warfarin-binding site of HSA was observed in TR-HSA. The reversible binding of F or naproxen to the warfarin-binding site of TR-HSA was less than to that of native HSA. These results indicate the photoactivated F was covalently bound to the warfarin-binding site of HSA. The covalent binding of Fgnd, which is also reversibly bound to the wafarin-binding site of HSA, was also induced by exposure to sunlight. Fgnd was more photoactive than F, indicating that F could be activated by glucuronidation to become a more photoactive compound.
Conclusions
The reactivity of photoactivated F and Fgnd to HSA and/or to other endogenous compounds may cause the phototoxic blisters that result at high F dosage.
Keywords: albumin, frusemide glucuronide, frusemide, photoallergy, phototoxicity
Introduction
Frusemide (F, 4-chloro-N-furfuryl-5-sulphamoylanthranilic acid)) has been widely used as a diuretic [1]. However, phototoxic blisters from high frusemide dosage have been reported [2, 3], but the reason for these has remained unclear. F can be photodegraded [4–6]. F in methanol is degraded primarily to N-furfuryl-5-sulphamoylanthranilic acid by photoreduction and 4-chloro-5-sulphamoyl-anthranilic acid (saluamine) by photohydrolysis when irradiated with 365 nm u.v. light [4]. F in aqueous solution [5, 6] is photodegraded to form 4-hydroxyl-N-furfuryl-5-sulphamoylanthranilic acid (4HFSA) by displacement of a chlorine with a hydroxyl group (Figure 1).
Figure 1.
Pathways of photodegradation and glucuronidation of F.
Recently, Selvaag & Thune reported that photodegradation products of sulphonamide-derived diuretics such as frusemide were not phototoxic in vitro [7]. Most of the F and frusemide glucuronide (Fgnd, metabolite of F) in blood are reversibly bound to the warfarin-binding site of human serum albumin (HSA) [8, 9]. Furthermore, we found that Fgnd in aqueous solution is also photodegraded to form the glucuronide of 4HFSA (4HFSAgnd) [6] (Figure 1). Therefore, in this report, we studied the effect of irradiation on F and Fgnd in the presence of human serum albumin (HSA) in terms of their reactivity with HSA, the site and the manner of their covalent binding to HSA, and the influence on reversible binding of drugs to the modified HSA after such treatment.
Methods
Reagents
F, indole-3-propionic acid (as internal standard for the h.p.l.c. assay of F) and HSA (essentially fatty acid free) were obtained from Sigma Chemical Co. Ltd (St Louis, MO). Naproxen (Nap) and indoprofen (as internal standard for the h.p.l.c. assay of Nap) were obtained from Syntex Labs., Inc. (Palo Alto, CA). Acetonitrile (h.p.l.c. grade) was obtained from J. T. Baker Chemical Co. (Philipsburg, NJ). Fgnd was prepared by the method reported previously [10]. All other chemicals were of reagent grade. All experiments were performed in a room which excluded sunlight, under fluorescent room lights.
Irradiation experiments
F, HSA, or a mixture of F and HSA, dissolved in 0.15 m phosphate buffer (pH 7.4) (HSA concentration, 100 or 500 μm; mole ratio of F to HSA, 1 or 5) was exposed to light (366 nm lamp, BLAK-RAY, VWR Scientific, San Francisco, CA), at a distance of 15 cm between the lamp and sample. Control sample (CONT) was prepared by mixing separately irradiated solutions of F and HSA. The experimental sample, which was obtained by irradiating a mixture of F and HSA, is referred to as SAMP. All solutions and mixtures were placed in glass tubes in a water bath to prevent an increase of the temperature of the samples.
Determination of photodegradation product of F
CONT or SAMP was treated with a volume of acetonitrile equal to that of the solution of CONT or SAMP. After precipitation of HSA by centrifugation of the mixture, the resultant supernatant (50 μl) was injected into the h.p.l.c. system as described below. The flow rate of the mobile phase (20% acetonitrile, 0.1 m ammonium acetate in water, pH 3.5 or 20% acetonitrile, 0.05% phosphoric acid in water, pH 3.5) was 1.3 ml min−1.
Effect of covalent binding of F on the fluorescence spectrum of HSA
Acetone (5 ml) was added into 0.1 ml of SAMP (500 μm HSA, 500 μm F) or HSA solution (500 μm). HSA was precipitated by centrifugation, and again 5 ml of acetone was added to the precipitate and the mixture was again centrifuged. This procedure was done three times to wash HSA in order to remove the compounds which were not covalently bound to HSA. After drying, the washed HSA was dissolved in 5 ml of 0.15 m potassium phosphate buffer (pH 7.4) for analysis by fluorescence spectrophotometry. The washed HSA which was derived from SAMP is referred to as TR-HSA. Fluorescence spectra were obtained by the method described above.
Determination of the compound covalently bound to HSA
The compound bound to HSA was recovered by the following method. Three ml of acetonitrile/ethanol (2:1) was added to 0.5 ml of CONT, SAMP or irradiated HSA solution. After centrifugation, the precipitated HSA was washed three times by addition of 1 ml of methanol/ether (3:1) and subsequent centrifugation. The washed HSA was hydrolysed with 0.3 ml of 0.25 or 0.1 m KOH (80° C, 30 min) after drying by nitrogen gas. The resulting solution was mixed with 20 μl of 21.75% phosphoric acid, and 100 μl of the mixture was injected into the h.p.l.c. system as described above.
Reversible binding of F and Nap to HSA
Reversible binding of F and Nap to HSA were examined using an ultrafiltration method with solutions of CONT and SAMP (500 μm HSA, 500 μm F) after irradiation for 8 h, under which condition F was completely degraded. Into 1 ml of five-fold diluted solution (100 μm HSA) of CONT or SAMP, F or Nap solution was added to give a concentration of 100–500 μm of F or Nap. The resulting solution was centrifuged using Amicon Centrifree (Amicon Division, W.R. Grace & Co, Danvers, MA) at 400 g for 10 min at room temperature after sampling 50 μl of the solution for determination of the concentration of F or Nap. Fifty microliters of filtrated solution was then sampled for determination of the concentration of free F or Nap. The sampled aliquots (50 μl) were mixed with acetonitrile (50 μl) containing internal standard (50 mm indol-3-propionic acid or 500 μm indoprofen), and the supernatant obtained after precipitation of HSA by centrifugation of the mixture was injected into the h.p.l.c. system.
The h.p.l.c. conditions for the F assay were as follows. An Altex ODS column (4.6 mm×25 cm, Beckman Instruments Inc., Berkeley, CA) was used. The mobile phase consisted of 30% acetonitrile, 0.05% phosphoric acid in water, pH 3.0. The flow rate was 1.5 ml min−1. Wavelengths of excitation and emission of fluorescence detector were set at 345 and 415 nm, respectively. The rest of the h.p.l.c. conditions was the same as described above. The h.p.l.c. conditions for the assay of Nap were as follows. The mobile phase consisted of 50% methanol, 0.3% acetic acid, pH 5.0. The flow rate was 1.5 ml min−1. An Altex Ultrasphere-OCTYL, 5 μm particle size column (4.6 mm×15 cm, Beckman Instruments Inc.) was used. The wavelength of the u.v. detector was set at 234 nm. The rest of the h.p.l.c. conditions was the same as described above.
Kinetic simulation of reversible binding of F and Nap to HSA
The reversible binding of F and Nap to HSA was simulated as follows. Since F was reported to bind to the wafarin-binding site [9], the reversible binding of F to HSA after irradiation in the absence of F was assumed to consist of a specific binding and nonspecific binding. Cb and Cf represent the concentrations of bound and free F, respectively. P is the concentration of HSA. n, Kd and Kp represent the number of binding sites, the dissociation constant and the partition coefficient for binding, respectively.
 |
The reversible binding of F to HSA after irradiation in the presence of F consists of the specific binding, with the number of binding sites (n−1), and nonspecific binding.
 |
The sites for binding of Nap to HSA after irradiation in the absence of F were assumed to consist of two classes of specific binding and nonspecific binding, because Nap was reported to bind to both the benzodiazepine-binding site and the warfarin-binding site [9]. n1 and n2 represent the numbers of primary (high affinity) and secondary (low affinity) specific binding sites, respectively. Kd1 and Kd2 represent the dissociation constants of the primary binding site and secondary binding site, respectively.
 |
The sites for binding of Nap to HSA after irradiation in the presence of F consist of a single class of specific binding (high affinity binding) sites and nonspecific binding sites.
 |
Covalent binding of F and Fgnd to HSA induced by photoactivation with sunlight
A mixture (0.5 ml) of 500 μm HSA and 500 μm F or Fgnd dissolved in 0.1 m phosphate buffer (pH 7.4) was exposed to sunlight for 30 min. An adduct to HSA was recovered as 4HFSA after alkali treatment. H.p.l.c. analysis was performed by the method described above. A standard sample of 4HFSA was prepared by purification of irradiated F solution with h.p.l.c., and identified by mass spectrometry.
H.p.l.c. system
The h.p.l.c. system consisted of a Beckman 110B M pump (Beckman Instruments Inc., Berkeley, CA), Kratos spectroflow 783 u.v. detector (Kratos Analytical, 170 Williams, Drive Ramsey, NJ), Shimadzu RF-530 fluorescence detector (Ex. 345 nm, Em. 415 nm) (Shimadzu Corporation, Kyoto, Japan) and Altex Ultrasphere-ODS column (4.6 mm×25 cm, 5 μm particle size) (Beckman Instruments Inc., Berkeley, CA).
Statistical analysis
All data are presented as mean±s.e.mean. The data of F, Fgnd and Nap were compared by paired or unpaired Student’s t-tests.
Results
Effect of HSA on the formation of photodegradation product
Figures 2a and b show the h.p.l.c. chromatograms of CONT and SAMP, which were obtained after 6 h irradiation of F in the absence and presence of HSA, respectively. 4HFSA was detected in both CONT and SAMP, but the amount of it in SAMP was much less than that in CONT.
Figure 2.
H.p.l.c. chromatograms of F solution after 6 h irradiation (a) in the absence of HSA and (b) in the presence of HSA.
Effect of covalent binding of F on fluorescence spectra of HSA
Fluorescence spectra of HSA in pH 7.4 buffer without irradiation in the absence of F (Figure 3a), without irradiation in the presence of F (Figure 3b) and after irradiation in the absence of F for 8 h (Figure 3c) show the same profiles when excited at the wavelengths of 280, 295 or 320 nm. On the other hand, the fluorescence spectra of HSA (TR-HSA) after irradiation in the presence of F for 8 h (Figure 3d) were completely different from those shown in Figure 3a-c. The fluorescence resulting from excitation of tyrosine and tryptophan residues of HSA at 280 nm was decreased in intensity and changed in shape by irradiation in the presence of F (Figure 3d). The fluorescence resulting from excitation of the lone tryptophan residue (Trp 214) of HSA at 295 nm was decreased in intensity, but the maximum emission wavelength was not charged (Figure 3d). On the other hand, when excited at 320 nm, a new fluorescence spectrum was observed (Figure 3d).
Figure 3.
Comparison of fluorescence spectra of HSA obtained under various conditions. Conditions; (a) 0 h in the absence of F (b) 0 h in the presence of F (c) 8 h in the absence of F and (d) 8 h in the presence of F. Numbers on spectra indicate excitation wavelengths.
Determination of F covalently bound to HSA
The HSA adduct formed after irradiation in the presence of F was detected by h.p.l.c. The retention time of the compound which was recovered after alkali treatment (0.25 m KOH, 80° C, 30 min) of HSA from SAMP, was around 11 min (Figure 4b), which matched the retention time of 4HFSA (Figure 2). On the other hand, its peak was not observed in the chromatogram of the HSA after irradiation in the absence of F or the chromatogram of HSA from CONT (Figure 4a and c).
Figure 4.
H.p.l.c. chromatograms of constituents released from HSA by alkali treatment. (a) HSA irradiated in the absence of F (b) HSA irradiated in the presence of F (c) HSA not irradiated in the presence of F.
Effect of covalent binding of F to HSA on reversible binding of F
The Rosenthal plots [11] of the reversible binding of F to HSA after irradiation in the absence and presence of F are shown in Figure 5a. This shows that reversible binding of F to HSA after irradiation in the absence of F (Figure 5a) consisted of a single class of specific binding and nonspecific binding, because the plots did not form straight lines. On the other hand, the values of Cb/Cf of the reversible binding of F at 100 and 200 μm to the HSA, which were obtained after irradiation in the presence of F, were significantly (P<0.05) lower than those of HSA, which were obtained after irradiation in the absence of F. That is, a significant difference was observed in the region describing the specific binding at bound concentrations of less than 200 μm, while the slope of plots in the region of bound concentrations of less than 200 m were similar to those in the absence of F. These results indicate that the number of specific binding sites was decreased after irradiation in the presence of F, and the F-binding ability of HSA after irradiation in the presence of F was lower than that in the absence of F.
Figure 5.
Rosenthal plots of the reversible binding of F (a) and Nap (b) to HSA after irradiation in the presence and absence of F for 8 h. Closed squares and open squares indicate the values in the absence of F and the values in the presence of F, respectively. Lines represent the simulation data.
The simulation data assuming that the number of specific binding sites (n)=3, Kd=40 μm and Kp=0.0055 for the reversible binding of F to HSA after irradiation in the absence of F and the number of specific binding sites (n−1)=2, Kd=40 μm and Kp=0.0055 for the reversible binding of F to HSA after irradiation in the presence of F, were described with curves in Figure 5a. These simulation data, which assume that the number of specific binding sites for F on HSA was decreased from 3 to 2 by irradiation in the presence of F, were well fitted to the experimental data, indicating that the photodegradation compound is bound to the wafarin-binding site on HSA.
Effect of covalent binding of F to HSA on reversible binding of Nap
The Rosenthal plots of the reversible binding of Nap to HSA after irradiation in the absence and presence of F were nonlinear, as seen in the Rosenthal plot of the reversible binding of F (Figure 5b). The values of Cb/Cf for the reversible binding of Nap at 100 and 200 μm to HSA after irradiation in the presence of F were significantly (P<0.05) lower than those in the absence of F. The slopes of the Rosenthal plots in the region describing the specific binding at bound concentrations of less than 200 μm of the reversible binding of Nap to HSA after irradiation in the absence and presence of F were not the same. The slope of the Rosenthal plot of the reversible binding of Nap to HSA after irradiation in the presence of F was larger than that in the absence of F, indicating that HSA after irradiation in the presence of F still has the high affinity specific binding sites.
The simulation data assuming the parameters n1=1, Kd1=1, n2=2, Kd2=90 μm and Kp=0.0075 for the reversible binding of Nap to HSA after irradiation in the absence of F and n1=1, Kd1=1 and Kp=0.0075 for the binding in the presence of F were described with the curves in Figure 5b. These simulation data were well fitted to the experimental data.
Photodegradation of F and Fgnd by exposure to sunlight
Photodegradation of F and Fgnd in the presence of 500 μm HSA in buffer solution at pH 7.4 is shown in Figure 6. About 90% of F remained after 30 min of exposure to sunlight, whereas only about 10% of Fgnd remained.
Figure 6.
Photodegradation of F and Fgnd by sunlight in the presence of HSA.Conditions: photo-exposure time, 30 min; F and Fgnd concentrations, 500 μm; HSA concentration, 500 μm. Data represent mean±s.e.mean (n=3). An asterisk represents significant difference (P<0.01).
Covalent binding of F and Fgnd to HSA induced by exposure to sunlight
The amount of covalent binding of F to HSA induced by exposure to sunlight for 30 min was 0.346±0.078 mmol/mol HSA (Figure 7), whereas the amount of covalent binding of Fgnd to HSA (15.7±1.69 mmol/mol HSA) was much greater than that of F.
Figure 7.
Effect of sunlight on the covalent binding of F and Fgnd to HSA.Conditions: photo-exposure time, 30 min; F and Fgnd concentration, 500 μm; HSA concentration, 500 μm. Data represent mean±s.e.mean (n=3). An asterisk represents significant difference (P<0.01).
Discussion
The formation of 4HFSA (photodegradation product of F) [6] after irradiation in the presence of HSA was less than that in the absence of HSA (Figure 2). This suggests that F reacted with HSA instead of aqueous hydroxyl ion, resulting in the decrease of the formation of 4HFSA. A new fluorescence spectrum was detected in the TR-HSA when excited at 320 nm (Figure 3d). Furthermore, 4HFSA was released by alkali hydrolysis of TR-HSA (Figure 4). 4HFSA contains an oxygen atom, but F does not (Figure 1). Therefore, these results indicate that irradiation caused F to become covalently bound to HSA by way of an oxygen atom which was derived from an amino acid residue of HSA, and 4HFSA was recovered from TR-HSA.
The treatment of TR-HSA with weaker alkali (0.1 m KOH) also released 4HFSA which was bound to HSA (data not shown). Therefore, the bond between the bound compound and the amino acid residue of HSA seems to be a ester bond rather than a ether bond. Moore & Sithipitaks [4] proposed the nucleophilic attack of solvent to a cation radical formed by irradiation as a mechanism for the photodegradation of F in methanol. Therefore, these covalent binding reactions can probably be attributed to the nucleophilic attack of a oxygen atom of an amino residue of HSA to a cation radical of F formed by irradiation.
The fluorescence spectra due to tyrosine and tryptophan residues of HSA excited at 280 nm were decreased in intensity and changed in shape by irradiation in the presence of F (Figure 3d). This suggests that tyrosine and/or tryptophan residue(s) were modified by irradiation of HSA in the presence of F, or that either tyrosine or tryptophan residues were quenched by a compound bound close to some amino acid residue of HSA after irradiation in the presence of F. The intrinsic fluorescence spectrum due to the lone tryptophan residue (Trp 214) of HSA excited at 295 nm was decreased in intensity after irradiation in the presence of F. Since the fluorescence (emission) spectrum of HSA overlapped with the absorption spectrum of 4HFSA [6], this indicates that the fluorescence of Trp 214 was quenched by fluorescence energy transfer from the tryptophan to the photodegradation compound covalently bound to HSA. Furthermore, the lone tryptophan residue of HSA is near the wafarin-binding site [12]. Therefore, it seems likely that F was covalently bound to the warfarin-binding site or near the site.
This likelihood is supported by the study of the reversible binding of F and Nap, which are bound to the warfarin-binding site [9]. The reversible binding of F to the HSA in SAMP was less than that to the HSA in CONT. The Rosenthal plot of the reversible binding of F (Figure 5a) indicates a decrease of the number of specific binding sites caused by irradiation in the presence of F. Furthermore, the reversible binding of Nap to HSA was decreased after irradiation in the presence of F (Figure 5b). The Rosenthal plot of Nap binding indicates that the primary binding site (high affinity), which is the benzodiazepine-binding site on HSA [9], and nonspecific binding sites remained after irradiation of HSA in the presence of F. This conclusion is supported by the simulation data, which assumed that HSA after irradiation in the presence of F has specific primary (high affinity) binding sites and nonspecific binding sites for Nap. Therefore, it is considered that the covalent binding of F to the warfarin-binding site inhibited the reversible binding of F or Nap to the warfarin-binding site (Figure 8). The covalent binding of F to the warfarin-binding site seems reasonable, because most F was reversibly bound to the warfarin-binding site of HSA [8] when F was irradiated in the presence of HSA.
Figure 8.
Scheme showing the photoinduced covalent binding of F and Fgnd to HSA.
Chlorpromazine photoreacts with methanol, yielding promazine and 2-methoxypromazine [13]. When chlorpromazine was irradiated in the presence of a solution of protein or nucleic acids, it lost chlorine to bind to the biopolymer [13]. Davies et al. [14] suggested that the promazine radical may be the reactive species that generates the antigen in vivo. The covalent binding of F to HSA was induced by exposure to sunlight as well (Figure 7). Furthermore, the covalent binding of Fgnd (F metabolite), which is also reversibly bound to the wafarin-binding site of HSA [8], was induced by exposure to sunlight, and was much greater than the binding of F (Figure 7). The covalent binding of F to HSA after irradiation suggests the possibility of a photoallergic response or toxicity in vivo after F administration and sunlight exposure [2, 3]. The reaction of F and Fgnd with other endogenous compounds may also cause phototoxicity. Furthermore, Fgnd was more photoactive than F, indicating that F could be activated by glucuronidation to become a more photoactive compound. Thus, phototoxic studies must be performed not only on drugs themselves but also on their metabolites.
Photodegradation of Nap [15] and clinical phototoxic responses of naproxen [16] have been reported. Results similar to those shown in Figure 2 were observed in experiments studying irradiation of Nap in the presence of HSA (data not shown). That is, the two main photodegradation products of Nap were formed much less in the presence of HSA than in the absence of HSA, suggesting that photoactivated Nap is bound to HSA.
The reactivity of photoactivated F shown in this study may be applicable to photoaffinity labelling studies. Since F is often used as an inhibitor for membrane transport study of organs such as liver [17], F or its derivatives may be able to be used as photoaffinity labelling reagents for transport carrier proteins.
In summary, photoactivated F was covalently bound to some amino acids of HSA which is located in or near the warfarin-binding site and the lone Trp 214. The adduct was formed by displacement of the chlorine of F with a hydroxyl group of the amino acid, and was released from the HSA by alkali treatment. Fgnd was also covalently bound to HSA by irradiation, and to a much greater extent than was F. This may be the cause of the phototoxic blisters formed at high F dosage.
References
- 1.Weiner IM, Mudge GH. Diuretics and other agents employed in the mobilization of edema fluid. In: Gilman AG, Goodman LS, Rall TW, Murad F, editors. The Pharmacological Basis of Therapeutics. Seventh. New York: Macmillan Publishing Company; 1985. pp. 887–907. [Google Scholar]
- 2.Burry JN, Lawrence JR. Phototoxic blisters from high frusemide dosage. Br J Dermatol. 1976;94:495–499. doi: 10.1111/j.1365-2133.1976.tb05136.x. [DOI] [PubMed] [Google Scholar]
- 3.Heydenreich G, Pindborg T, Schmidt H. Bullous dermatosis among patients with chronic renal failure on high dose frusemide. Acta Med Scand. 1977;202:61–64. doi: 10.1111/j.0954-6820.1977.tb16784.x. [DOI] [PubMed] [Google Scholar]
- 4.Moore DE, Sithipitaks VJ. Photolytic degradation of frusemide. J Pharm Pharmacol. 1983;35:489–493. doi: 10.1111/j.2042-7158.1983.tb04816.x. [DOI] [PubMed] [Google Scholar]
- 5.Bundgaard H, Norgaard T, Nielsen NM. Photodegradation and hydrolysis of furosemide and furosemide esters in aqueous solutions. Int J Pharm. 1988;42:217–224. [Google Scholar]
- 6.Mizuma T, Benet LZ, Lin ET. High performance liquid chromatographic determination and identification of photodegradation and acyl migration products of furosemide 1-O-acyl glucuronide. J Chromatogr B. 1998;718:153–162. doi: 10.1016/s0378-4347(98)00357-0. [DOI] [PubMed] [Google Scholar]
- 7.Selvaag E, Thune P. Photodegradation products of sulphonamide-derived oral antidiabetics and diuretics are not phototoxic in vitro. Photodermatol Photoimmunol Photomed. 1996;12:79–83. doi: 10.1111/j.1600-0781.1996.tb00179.x. [DOI] [PubMed] [Google Scholar]
- 8.Mizuma T, Benet LZ, Lin ET. Abstract of the Fourth Japanese American Conference on Pharmacokinetics and Biopharmaceutics. San Francisco; 1988. Isomerization and interaction of furosemide 1-O-acyl glucuronide with albumin. [Google Scholar]
- 9.Sjoholm I, Ekman B, Kober A, Ljungstedt-Pahlman I, Seiving B, Sjodin T. Binding of drugs to human serum albumin: XI. The specificity of three binding site as studied with albumin immobilized in microparticles. Mol Pharmacol. 1979;16:767–777. [PubMed] [Google Scholar]
- 10.Mizuma T, Benet LZ, Lin ET. Preparative chromatography of furosemide 1-O-acyl glucuronide from urine using micronized Amberlite XAD-2 and its application to other 1-O-acyl glucuronides. Prep Biochem Biotechnol. 1998;28:37–47. doi: 10.1080/10826069808010125. [DOI] [PubMed] [Google Scholar]
- 11.Rosenthal HE. A graphic method for the determination and presentation of binding parameters in a complex system. Anal Biochem. 1967;20:525–532. doi: 10.1016/0003-2697(67)90297-7. [DOI] [PubMed] [Google Scholar]
- 12.Fehske KJ, Muller WE, Wollert U. The lone tryptophan residue of human serum albumin as part of the specific warfarin binding site. Binding of dicoumarol to the warfarin, indole and benzodiazepine binding sites. Mol Pharmacol. 1979;16:778–789. [PubMed] [Google Scholar]
- 13.Rosenthal I, Ben-Hur E, Prager A, Riklis E. Photochemical reactions of chlorpromazine; chemical and biochemical implications. Photochem Photobiol. 1978;28:591–594. doi: 10.1111/j.1751-1097.1978.tb06975.x. [DOI] [PubMed] [Google Scholar]
- 14.Davies AK, Navartnam P, Philips GO. Photochemistry of chlorpromazine [2-chloro-N-(3-dimethylaminopropyl) phenothiazine] in propan-2-ol solution. J Chem Soc Perkin Trans II. 1976:25–29. [Google Scholar]
- 15.Moore DE, Chappuis PP. A comparative study of the photochemistry of the non-steroidal anti-inflammatory drugs, naproxen, benoxaprofen and indomethacin. Photochem Photobiol. 1988;47:173–180. doi: 10.1111/j.1751-1097.1988.tb02710.x. [DOI] [PubMed] [Google Scholar]
- 16.Diffey BL, Daymond TJ, Fairgreaves H. Phototoxic reactions to piroxicam, naproxen and tiaprofenic acid. Br J Rheumatol. 1983;22:239–242. doi: 10.1093/rheumatology/22.4.239. [DOI] [PubMed] [Google Scholar]
- 17.Blitzer BL, Terzakis C, Scott KA. Hydroxyl/bile acid exchange. A new mechanism for the uphill transport of cholate by basolateral liver plasma membrane vesicles. J Biol Chem. 1986;26:12042–12046. [PubMed] [Google Scholar]