| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 279, Issue 29, 30009-30020, July 16, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Pharmacology and Toxicology, College of Pharmacy, Arizona Cancer Center, University of Arizona, Tucson, Arizona 85724
Received for publication, April 20, 2004 , and in revised form, May 3, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Preparation of a Covalent Bovine Serum Albumin-Vitamin B6 Conjugate (Pyridoxylated BSA, BSA-B6)—BSA-B6 was synthesized by reductive coupling of pyridoxal with protein lysine residues based on NaCNBH3 reduction of the initially formed Schiff base as described earlier (19). The reaction mixture contained BSA (350 mg), NaCNBH3 (58 mg), and pyridoxal (64 mg) in a total volume of 1.5 ml of 0.25 M phosphate buffer, pH 7.4. The reaction proceeded at 37 °C in the dark overnight. The reaction was terminated by acidification, followed by neutralization and extensive dialysis against water at 4 °C for 48 h under light exclusion. After lyophilization, the protein adduct, called BSA-B6, was characterized by MALDI-TOF mass spectrometry and fluorescence and UV spectroscopy.
Irradiation with Solar Simulated Light (SSL) and UVA—A KW large area light source solar simulator, model 91293, from Oriel Corp. (Stratford, CT) was used, equipped with a 1000-watt xenon arc lamp power supply, model 68920, and a VIS-IR band pass blocking filter plus either an atmospheric attenuation filter (output 290–400 nm plus residual 650–800 nm for solar simulated light) or UVB and UVC blocking filter (output 320–400 nm plus residual 650–800 nm, for UVA), respectively. The output was quantified using a dosimeter from International Light Inc. (Newburyport, MA), model IL1700, with an SED240 detector for UVB (range 265–310 nm, peak 285 nm) or a SED033 detector for UVA (range 315–390 nm, peak 365 nm) at a distance of 365 mm from the source, which was used for all experiments. At 365 mm from the source, SSL dose was 7.63 mJ cm–2 s–1 UVA and 0.40 mJ cm–2 s–1 UVB radiation. Using a UVB/C-blocking filter, the dose at 365 mm from the source was 5.39 mJ cm–2 s–1 UVA radiation with a residual UVB dose of 3.16 µJ cm–2 s–1.
Cell Culture—The established cell line of human epidermal keratinocytes (HaCaT cells), a gift from Dr. Norbert Fusenig (German Cancer Research Center, Heidelberg, Germany), and human dermal fibroblasts (CF-3 cells), a gift from Dr. Robert Dell'Orco (Noble Center for Biomedical Research, Oklahoma City, OK) were routinely cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and kept in a humidified atmosphere containing 5% CO2 at 37 °C (17).
Assay for Photosensitized Suppression of Skin Cell Proliferation— Cells were seeded at 5 x 104 cells/dish on 35-mm dishes. After 24 h, cells were washed, placed in Hanks' balanced salt solution (HBSS), and exposed to the combined or isolated action of photodynamic test compound and irradiation (UVA or SSL). Following a 30-min incubation after irradiation, the exposure medium was removed and replaced with fresh culture medium. Cell number was determined 72 h later, and proliferation was compared with cells that received mock irradiation in HBSS.
Quantification of H2O2 Formation—H2O2 formed upon photosensitization was quantified according to a standard procedure using the ferrous iron-xylenol orange assay as reported previously (13).
Detection of Intracellular Oxidative Stress by Flow Cytometric Analysis—Intracellular levels of peroxides were analyzed by flow cytometry using dihydrorhodamine 123 (DHR 123) as a specific fluorescent dye probe (17, 20, 21). To avoid direct photooxidation of the dye probe, cells were first treated with UV irradiation and sensitizer and then loaded with the indicator dye under light exclusion.
Apoptosis Analysis—Induction of cell death was confirmed by annexin V-fluorescein isothiocyanate/propidium iodide (PI) dual staining of cells followed by flow cytometric analysis. Cells (200,000) were seeded on 35-mm dishes and received photosensitization 24 h later. Cells were harvested at various time points after treatment, and cell staining was performed using an apoptosis detection kit according to the manufacturer's specifications (APO-AF; Sigma).
Cell Cycle Analysis—Cells were seeded at 2 x 105/dish on 35-mm culture dishes (Sarstedt) and left overnight to attach. After irradiation in the presence or absence of test compound, cells were washed twice with HBSS, and fresh culture medium was added. After 24 or 48 h, cells were harvested by trypsinization, resuspended in 200 µl of PBS, and placed on ice. After the addition of 2 ml of 70% (v/v) ethanol, 30% (v/v) PBS, cells were incubated for 30 min on ice. The fixed cells were pelleted by centrifugation; resuspended in 800 µl of PBS, 100 µl of ribonuclease A (1 mg/ml PBS), and 100 µl of propidium iodide (400 µg/ml PBS); and incubated for 30 min in the dark at 37 °C. Cellular DNA content was determined by flow cytometry using the ModFit LT software, version 3.0 (Verity, Topsham, ME).
p38 Mitogen-activated Protein (MAP) Kinase Western Analysis—One day before treatment, 200,000 cells were seeded on 35-mm dishes. Cells were washed twice 24 h later with HBSS, followed by the addition of 1 ml of HBSS and exposure to increasing doses of UVA (up to 9.9 J/cm2) in the absence or presence of 100 µM NE-3HP. After irradiation or mock treatment, cells were incubated for 30 min (37 °C, 5% CO2) and then washed with PBS, lysed in 1x SDS-PAGE sample buffer (200 µl, 0.375 M Tris-HCl, pH 6.8, 50% glycerol, 10% SDS, 5% 
-mercaptoethanol, 0.25% bromphenol blue), heated for 5 min at 95 °C, and placed on ice. Samples (23 µl, containing 
45 µg of total protein as determined by the BCA assay) were separated by 12% SDS-PAGE followed by immediate transfer to polyvinylidene difluoride membranes (Immobilon; Millipore Corp., Bedford, MA). Equal protein loading and transfer was examined by reversible Ponceau staining, and the membrane was then used for p38 immunostaining as published recently (22).
Sensitization of Protein Photocross-linking and Peptide Photo-oxidation—Ribonuclease A (RNase A, 10 mg/ml PBS) was irradiated with solar simulated light (13.8 J/cm2 UVA and 0.72 J/cm2 UVB) in the absence or presence of various hydroxypyridine derivatives (500 µM each, 200-µl total reaction volume). Protein oligomerization as a result of sensitized photocross-linking was visualized by 15% SDS-PAGE followed by Coomassie staining and densitometric analysis using an Eagle Eye digital camera (Stratagene, La Jolla, CA). Melittin (1 mg/ml PBS) was irradiated with solar simulated light (13.8 J/cm2 UVA and 0.72 J/cm2 UVB) in the absence or presence of various hydroxypyridine derivatives (500 µM each, 200-µl total reaction volume) followed by mass spectrometry.
Spectroscopy—UV spectra were recorded using a Cary 100 Bio UV-visible spectrophotometer from Varian, Inc. (Palo Alto, CA). Fluorescence spectra were recorded using a Spectramax Gemini XS (Molecular Devices, Inc., Sunnyvale, CA) 96-well microtiter plate reader.
Mass Spectrometry—Mass spectrometry was performed using a Bruker Reflex III MALDI-TOF mass spectrometer equipped with a nitrogen laser (337 nm). Spectra were recorded in positive ion mode in linear configuration using 
-cyano-4-hydroxycinnamic acid as matrix.
Statistical Analysis—The results are presented as means ± S.D. of at least three independent experiments. They were analyzed using the two-sided Student's t test (*, p < 0.05; **, p < 0.01; ***, p < 0.001).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
60%). A suppression of cell proliferation by almost 45% was observed, when unirradiated cells were exposed to 6 µM H2O2, an amount equal to the concentration formed upon UVA irradiation of 500 µM pyridinoline, as quantified in Fig. 2B. Consistent with photosensitized formation of H2O2 as the mechanism of inhibition of proliferation by UV-irradiated pyridinoline, complete suppression of photosensitization was achieved when the antioxidants catalase or D-penicillamine were added before irradiation, whereas no protection was achieved using the singlet oxygen quencher sodium azide. No photosensitization of ROS formation or inhibition of skin cell proliferation was observed when cells were irradiated in the presence of desmosine. Next, we tested pyridoxine (PN; see Fig. 1), another endogenous hydroxypyridine derivative and structural homologue of pyridinoline (23), for photosensitization of H2O2 production and inhibition of skin cell proliferation. Pyridoxine was a sensitizer of UVA-driven H2O2 formation equally effective as pyridinoline (Fig. 2B), but sensitized suppression of fibroblast proliferation exceeded the effect of H2O2 formed during irradiation (Fig. 2B). In contrast to pyridinoline sensitization, catalase treatment only weakly reversed pyridoxine sensitization, which suggests a mechanism of pyridoxine phototoxicity operating in addition to sensitization of H2O2 formation. Among various antioxidants including superoxide dismutase, mannitol, and deferoxamine mesylate (data not shown), only penicillamine treatment effectively antagonized pyridoxine-photosensitization (Fig. 2A). 1O2 involvement in pyridoxine photosensitization was excluded based on the observation that the presence of the singlet oxygen quenchers NaN3 and DABCO (data not shown) during irradiation was not protective, and irradiation in deuterated PBS did not enhance photosensitization (data not shown). From these initial observations, we conclude that the hydroxypyridine derivatives pyridinoline and pyridoxine are UVA photosensitizers of light-driven ROS formation and that inhibition of skin cell proliferation in the case of pyridoxine depends only partially on ROS formation.
|
|
80% proliferation of unirradiated controls; data not shown), was calculated by extrapolation from proliferation-inhibition curves (shown in Fig. 3A) for pyridinoline (632 ± 53 µM), pyridoxine (104 ± 17 µM), pyridoxamine (32 ± 7 µM), and pyridoxal (5 ± 1 µM). At concentrations approximately 5 times the IC50 for suppression of proliferation, pronounced induction of skin cell apoptosis was observed after combined exposure to B6 vitamers and UVA irradiation as detected by annexin V-fluorescein isothiocyanate/PI staining followed by flow cytometric analysis. Photosensitization of cell kill was particularly pronounced with pyridoxal as shown in Fig. 3B, with 35% of gated cells in either early or late apoptosis/necrosis at 24 h after irradiation. This induction of cell death was suppressed when the sensitization was performed in the presence of the thiol-antioxidant and free radical scavenger penicillamine. As observed above in Fig. 2A, NaN3 did not show photoprotection, suggesting that singlet oxygen is not involved in pyridoxal-photosensitized induction of cellular apoptosis (data not shown).
|
-amino groups of protein lysine residues in bovine serum albumin (BSA) performed after initial Schiff base formation (19). The covalent conjugate (BSA-B6, bearing pyridoxylated lysine residues as shown in Fig. 1) was characterized by MALDI-TOF mass spectrometry. A mass difference 
m (BSA-B6 – BSA; 66,965 – 65,849) of 1100 Da indicated the incorporation of approximately seven pyridoxine moieties per BSA molecule. No residual unmodified BSA or species of intermediate weight were detected in the mass spectrum of BSA-B6 (data not shown). The UV spectrum of the BSA-B6 conjugate exhibited a novel absorption maximum at 335 nm (Fig. 3C), and fluorescence excitation at 335 nm led to strong emission around 400 nm (Fig. 3D), consistent with B6-fluorophore incorporation into the protein. BSA-B6 was devoid of any dark toxicity, since exposure of skin fibroblasts to BSA-B6 (5 mg/ml PBS) in the absence of UVA irradiation did not inhibit proliferation as shown in Fig. 3E. In contrast, pronounced inhibition of cell proliferation resulted from exposure to the combined action of UVA and BSA-B6, which was not observed upon irradiation in the presence of unmodified or NaCNBH3-reduced BSA. Based on these data, we conclude that unphosphorylated or 5'-phosphorylated B6 vitamers are UVA photosensitizers and that protein-bound pyridoxine retains phototoxicity.
Identification of the Minimum Phototoxic Chromophores Contained in Endogenous Hydroxypyridine Photosensitizers— In an attempt to identify the minimum chromophore responsible for the photosensitizer activity of the endogenous hydroxypyridine derivatives pyridinoline and B6 vitamers, the hydroxypyridine (HP) isomers 2-HP, 3-HP, and 4-HP and the N-alkyl-3-HP derivative and pyridinoline analogue (25) NE-3HP (100 µM each; formulas given in Fig. 1) were tested for photosensitized suppression of HaCaT keratinocyte proliferation by low doses of UVA (see Fig. 4A). None of the test compounds exhibited significant dark toxicity, and 3 days after combined treatment with test compound and UVA, only NE-3HP strongly suppressed cell proliferation. Cell counts obtained from samples treated with light and NE-3HP were below seeding density, suggesting cell depletion by photosensitized induction of cell death. Induction of apoptosis by the combined action of NE-3HP and UVA was confirmed by annexin V-fluorescein isothiocyanate/PI staining and flow cytometric analysis performed over 24 h after cell treatment (Fig. 4B). Cells in early (annexin V+/PI–) and late (annexin V+/PI+) apoptosis were observed starting 6 h after treatment (data not shown), and 24 h later, almost all cells were in late apoptosis/necrosis. These changes were not observed after administration of test compound or light alone. Next, the experiment was repeated using SSL instead of UVA to elucidate the possible contribution of UVB wavelengths to UV photoactivation of the test compounds as shown in Fig. 4C. With SSL, NE-3HP again induced almost complete inhibition of proliferation of HaCaT keratinocytes, but in this case 3-HP also was cytostatic, indicating UVB activation of 3-HP phototoxicity. The differential range of UV wavelengths effective for the induction of 3-HP and NE-3HP phototoxicity clearly correlate with the distinct UV absorption and fluorescence characteristics of these chromophores; 3-HP is a nonfluorescent compound characterized by broad UV absorption with maxima centered at 
(): 245 nm (2477), 276 nm (2048), and 312 nm (4435). NE-3HP shows UV absorption maxima centered at () (248 nm (7506) and 319 nm (5243)) and displays intense fluorescence upon excitation at UVB and UVA I (320–340-nm) wavelengths due to broad excitation and emission maxima (ex/em = 305/395 nm) very similar to its structural analogue pyridinoline (25, 26). In contrast to the rapid photodegradation of pyridinoline observed with SSL or UVA irradiation described previously (17), prolonged exposure to UVA induced only a minor degradation of NE-3HP (<5%) as measured by fluorescence and UV spectroscopy (data not shown). To gain further insight into the mechanism of antiproliferative action of 3-HP sensitization, cell cycle analyses were performed 24 h after photosensitization using the combined action of 100 µM 3-HP and SSL (2.3 J/cm2 UVA, 120 mJ/cm2 UVB) as shown in Fig. 4D. The pronounced accumulation of cells with 4n DNA content and depletion of cells with 2n DNA content is consistent with the sensitized induction of a G2/M block. Treatment with the sensitizer alone did not induce any alterations of the cell cycle, but the isolated action of SSL in the absence of sensitizer induced a moderate accumulation of cells in S and G2/M phase consistent with the significant suppression of HaCaT cell proliferation by 20% for SSL irradiation alone as shown in Fig. 4C. Next, a comparative dose response was established for the sensitized inhibition of proliferation by 3-HP and NE-3HP. At an SSL dose of 2.3 J cm–2 UVA plus 0.12 J cm–2 UVB, the sensitizer concentration that caused 50% suppression of cell proliferation (IC50± S.D., n = 3) observed after light treatment alone (80% proliferation of unirradiated controls) was calculated by extrapolation from proliferation-inhibition curves (data not shown) for 3-HP (36 ± 11 µM) and NE-3HP (4 ± 1 µM). Based on this quantitative structure-activity relationship study of skin cell photosensitization by hydroxypyridine derivatives, 3-hydroxypyridine was identified as the minimum UVB sensitizer chromophore, and N-alkyl-3-hydroxypyridinium cation was identified as the minimum UVA sensitizer chromophore contained in the endogenous hydroxypyridine photosensitizers pyridoxine, pyridoxamine, pyridoxal, and pyridinoline, respectively.
|
3-fold increase over base-line fluorescence observed after loading the cells with the redox dye as shown in Fig. 5A, which is consistent with UVA-induced intracellular oxidative stress (21). Exposure to sensitizer alone did not induce any enhancement of base-line fluorescence. When cells were exposed to the combined action of UVA and pyridoxine, an additional 2-fold increase of rhodamine fluorescence intensity was observed, demonstrating the enhancement of UVA-induced photooxidative stress by photosensitization. Similar results were obtained when cells were exposed to the combined action of UVA (3.3 J cm–2) and 3-hydroxypyridine (100 µM), pyridinoline (500 µM), or NE-3-HP (100 µM) (data not shown). These data support the hypothesis that photosensitization of skin cells by 3-HP derivatives enhances intracellular photooxidative stress with formation of reactive species of sufficient longevity, such as protein peroxides (28), capable of oxidizing DHR 123 during cell loading after irradiation.
|
To further examine the involvement of oxidative pathways in the photosensitization of human skin cells by 3-HP derivatives, antioxidant modulation of sensitized inhibition of skin cell proliferation was examined as depicted in Fig. 5C. As observed earlier with pyridoxine, D-penicillamine fully reversed the sensitized inhibition of skin cell proliferation resulting from combined treatment with SSL and 3-HP or NE-3HP, when present during irradiation, but was completely ineffective when added immediately after irradiation. Other antioxidants such as deferoxamine mesylate, mannitol, superoxide dismutase, catalase, and DABCO did not suppress the sensitized inactivation of HaCaT keratinocytes by 3-HP derivatives. Inhibition of proliferation was not enhanced when the irradiation was performed in deuterated PBS (data not shown), providing further evidence that singlet oxygen is not involved in the phototoxic action of 3-HP derivatives on human skin cells.
Protein Damage as a Result of Hydroxypyridine Photosensitization—Cellular photosensitization is thought to be triggered by photooxidation of biological macromolecules (28, 31). Photosensitization of protein damage by 3-HP derivatives was examined using a ribonuclease A (RNase A) photocross-linking assay. RNase A was selected as a model target because it does not contain tryptophan residues, thereby excluding effects of this endogenous UV sensitizer amino acid on photocross-linking (13). RNase A (monomer; 13,700 Da) was irradiated with SSL in the presence or absence of various hydroxypyridine derivatives, and covalent protein oligomerization (dimer, 28,000 Da; trimer, 42,000 Da; etc.) was examined using reducing SDS-PAGE analysis. As shown in Fig. 6, A and B, sensitized photocross-linking occurred in the presence of 3-HP, NE-3HP, pyridoxine, pyridoxamine, and pyridoxal (data not shown), whereas SSL irradiation in the presence of 2-HP, 4-HP, pyridinoline, and desmosine was ineffective in generating RNase oligomers. Among the test compounds, sensitization by pyridoxine was most pronounced, yielding 35% RNase oligomerization. Photocross-linking sensitized by the minimum sensitizer chromophore 3-HP was examined in more detail. Sensitization occurred dose-dependently with regard to light dose (data not shown) and sensitizer concentration as shown in Fig. 6C. Antioxidant effects on pyridoxine-sensitized protein damage were examined next as shown in Fig. 6D. Antioxidant modulation of pyridoxine-sensitized protein cross-linking clearly paralleled antioxidant inhibition of cell proliferation observed earlier as shown in Fig. 2A. Among various antioxidants used to test for antioxidant modulation of sensitized protein damage, only the thiol antioxidant, copper ion chelator, and free radical quencher D-penicillamine (P) fully suppressed pyridoxine-sensitized photocross-linking. Hydroxyl radical scavenging using mannitol, iron ion chelation using deferoxamine mesylate, scavenging of H2O2 and superoxide radical anion using catalase, and superoxide dismutase, respectively, and the excited singlet state quencher potassium iodide (13) did not interfere with pyridoxine sensitization of protein photodamage. In addition, pyridoxine-sensitized photocross-linking was not enhanced when the reaction was performed in deuterated PBS, again suggesting that singlet oxygen was not involved (data not shown). Further evidence against the involvement of singlet oxygen in pyridoxine-sensitized RNase photo-oligomerization was provided by the studies with the singlet oxygen quencher 1,4-diazabicyclo[2.2.2]octane, which unexpectedly enhanced protein damage, compatible with facilitated electron transfer reactions in the presence of a DABCO radical cation as reported previously (32). Moreover, photocross-linking effectively proceeded when the irradiation was performed under argon, demonstrating that pyridoxine photosensitization of protein damage can occur by oxygen-independent reaction pathways (data not shown). Similar data were obtained when photocross-linking was sensitized by 3-HP or NE-3HP (data not shown). These data have led us to conclude that 3-HP derivatives are potent sensitizers of protein photocross-linking and that 3-HP-sensitization can proceed in the presence or absence of oxygen by unidentified reaction pathways compatible with type I photoreactions with initial formation of photoexcited states of the sensitizer followed by electron transfer reactions, either of which could effectively be quenched by thiol compounds.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In this study, biogenic 3-hydroxypyridine derivatives contained in human skin have been identified as a novel class of potential endogenous photosensitizers. Skin cell photosensitization by various endogenous hydroxypyridine chromophores contained in human skin is demonstrated in Figs. 2 and 3, and 3-HP and N-alkyl-3HP are identified as the minimum chromophores responsible for photosensitization toward SSL and UVA, respectively, as shown in Fig. 4. Photosensitization of cultured human skin cells by 3-HP derivatives resulted in dose-dependent inhibition of proliferation, cell cycle arrest, and induction of apoptosis. Multiple lines of experimental evidence for an involvement of photooxidative mechanisms in skin cell phototoxicity of 3-HP derivatives were obtained as presented in Fig. 5; photosensitization by 3-HP derivatives strongly potentiated UVA-induced intracellular peroxide formation and could be suppressed by thiol antioxidant intervention. UVA induction of p38 MAP kinase-dependent stress signaling, known to regulate AP-1-activation and matrix metalloproteinase expression in skin fibroblasts (27, 29), was strongly enhanced by 3-HP photosensitization. However, no specific macromolecular target of skin cell photosensitization by biogenic 3-HP derivatives was identified at this point, although photosensitization of protein and peptide damage was demonstrated in model systems as summarized in Figs. 6 and 7.
The mechanistic basis of hydroxypyridine phototoxicity remains incompletely understood at this point. Particularly, the complex photochemistry of B6 vitamers and their photoproducts has been studied in detail (35–38), but the involvement of specific photoproducts in skin cell photooxidative stress observed in this study remains to be elucidated. Obviously, photosensitization by pyridinoline fully depends on the light-driven formation of superoxide and H2O2 consistent with earlier observations (17), whereas photosensitization of ROS formation is not essential for skin cell phototoxicity of B6 vitamers, 3-HP, and NE-3HP, which was demonstrated by oxygen independence and a general lack of antioxidant suppression, except inhibition in the presence of penicillamine. Indirect experimental evidence in support of a predominant involvement of type I photosensitization mechanisms initiated by electron transfer reactions between photoexcited 3-HP, NE-3HP, or B6 vitamers and substrate molecules was obtained from (i) the unique inhibitory activity of the thiol-type free radical scavenger penicillamine on sensitized protein and skin cell damage, (ii) the complete ineffectiveness of other antioxidant treatments and anaerobic conditions to block sensitized protein cross-linking, and (iii) a lack of enhancement of photosensitization in deuterated buffer and the ineffectiveness of the singlet oxygen quenchers DABCO and NaN3 to protect cells or suppress protein damage (27, 39). Further evidence against involvement of singlet oxygen formation was based on the fact that pyridoxine and 3-HP are known singlet oxygen quenchers (37, 38). Obviously, the presence of a phenolic 3-OH substituent is an essential structural requirement for sensitizer activity of all tested pyridine derivatives, since desmosine, devoid of any hydroxyl substituent, and 2-HP and 4-HP, which occur predominantly as the tautomeric pyridone structures in aqueous solutions at neutral pH (40), display no photosensitizer activity. The phenolic character of 3-HP derivatives may be of crucial importance for the observed sensitization effects, since it is well documented that upon photoexcitation phenolic substances release electrons into aqueous solutions by electron ejection with formation of phenoxyl-type organic free radicals from the excited triplet state but conversely act as potent electron scavengers in the ground state (41). Moreover, single electron transfer reactions leading to the formation of 3-hydroxypyridinium (42) and pyridoxine (43) free radicals have been reported previously, and free radical polymerization of synthetic N-substituted 3-oxypyridinium betaines can be initiated by UV irradiation (44), consistent with a free radical mechanism of phototoxicity of biogenic 3-HP derivatives observed in this study. Obviously, the differences in mechanism and potency of photosensitization observed with various 3-HP derivatives are a consequence of substituent effects on hydroxypyridine photochemistry. The enhanced phototoxicity of pyridoxal as compared with the other B6-vitamers is consistent with the electron-withdrawing effects of the carbaldehyde substituent and facilitated hydrogen transfer, known to occur in aromatic compounds with a hydroxyl group in ortho-position to a carbonyl group (35). Likewise, the strong difference in photosensitization potency and mechanism between the similar fluorophores N-ethyl-3-HP and pyridinoline must be a consequence of pyridinium ring substitution by bulky, charged alkyl residues that might inhibit type I interaction with larger substrate molecules but allow interaction with molecular oxygen, leading to ROS formation. However, detailed photochemical studies employing laser flash photolysis and electron spin resonance techniques will be necessary to elucidate the excited state chemistry and free radical species involved in photosensitization by members of the 3-HP group of sensitizers. In addition, differential cellular uptake and metabolism probably will contribute to the remarkable differences with regard to phototoxicity and mechanism of sensitization observed with biogenic and synthetic 3-HP derivatives, which must be elucidated in the future.
The identification of 3-HP as a potent phototoxic chromophore contained in various human skin molecules must be considered in light of increasing evidence for the involvement of endogenous photosensitizers as key mediators of UVA-induced skin photoaging and photocarcinogenesis (1, 3, 4). In this study, photoactivation of physiologically relevant concentrations of 3-HP derivatives occurred by irradiation with doses of UVA and SSL well below the minimal erythemal threshold (45), equivalent to a short exposure of fair human skin to full spectrum solar UV irradiation (7). Currently, the occurrence of 3-HP chromophores is established in three groups of ubiquitous tissue biomolecules: (i) the B6 vitamers pyridoxine, pyridoxamine, pyridoxal, and their respective phosphorylated derivatives and protein conjugates; (ii) the ECM protein-associated cross-links pyridinoline and deoxypyridinoline; and (iii) various advanced glycation end products (AGEs) that accumulate on long lived proteins as a result of chronic carbonyl stress during chronological aging (46–48). (i) Increased photosensitivity is a known consequence of B6 overdosing in humans (49). Phototoxicity of UV-irradiated pyridoxamine was reported as early as 1947 by Shwartzman and Fisher (50), followed by other reports thereafter (36, 51). B6 vitamers therefore are important as endogenous skin photosensitizers with relevance to human skin in vivo, since human skin contains various B6 vitamer forms in significant amounts (100 nmol/g of protein (24)), with pyridoxal-5'-phosphate and pyridoxal being the predominant vitamers in vivo. Pyridoxal was clearly the most phototoxic vitamer tested in our assays with pronounced apoptogenicity toward skin fibroblasts after photoactivation at only moderate doses of UVA. In contrast to earlier observations (35), pyridoxal-5'-phosphate was as potent as pyridoxal in photosensitizer-induced skin cell apoptosis, consistent with 3-HP being the minimum chromophore required for sensitizer activity. Our results on B6 vitamer phototoxicity also point toward a potential risk of photosensitization associated with the use of high doses of pyridoxamine currently in clinical development as a therapeutic intervention for the inhibition of glycation-associated diabetic complications and hyperlipidemia (52). The phototoxic role of B6 vitamers in human skin as shown by our experiments strongly contrasts with but does not contradict a protective role of pyridoxine against singlet oxygen damage observed in a phytotoxic fungus (38). (ii) Skin collagen pyridinoline content is generally low (16 mmol/mol of collagen (53, 54)) but dramatically increases during conditions of wound healing, scar formation, and sclerotic disorganization as referenced in Ref. 17. It is tempting to speculate that a significant increase in pyridinoline and deoxypyridinoline content in skin collagen characteristic of scar tissue and sclerotic skin diseases introduces an endogenous UVA sensitizer that may contribute to the known predisposition of scar tissue toward photocarcinogenesis (55), a hypothesis to be explored in the future. (iii) AGEs, cross-link chromophores formed by nonenzymatic amino-carbonyl reactions between sugars and protein-bound amino groups (glycation), accumulate on skin ECM proteins during conditions of increased carbonyl stress, such as actinic (56) and chronological aging and diabetes (57). The phototoxic activity of AGEs extracted from skin collagen and lens crystallin is well documented (13, 58), and photosensitization of skin cell photooxidative stress results from UVA irradiation of AGEs (14, 59). Increasing experimental evidence indicates that 3-HP epitopes are formed during tissue glycation. N-Alkyl-3HP derivatives are formed during glycation and lipid peroxidation under physiological conditions (60). Importantly, AGE-fluorophores of the N-alkyl-3HP type have been isolated in significant amounts from human tissue, such as glycolaldehyde-pyridine from human atherosclerotic lesions (47). Moreover, Lys-hydroxytriosidines, another protein cross-link of the N-alkyl-3-HP type was isolated from human cornea collagen exposed to the artificial tanning agent dihydroxyacetone (48) and is therefore expected to form in appreciable amounts in human skin exposed to this compound. The phototoxicity of 3-HP derivatives as described in this study therefore raises the possibility that chemical tanning of human skin places a suspected photosensitizer in direct proximity to sensitive targets and adds to the increasing health concerns associated with the cosmetic use of dihydroxyacetone preparations (61). Experiments using isolated AGE products of the 3-HP type will allow testing of their phototoxicity and validation of a potential involvement in AGE photosensitization of human skin cells (14).
Future research will explore a functional involvement of the diverse members of the 3-HP group of endogenous photosensitizers in photodamage and carcinogenesis of UVA-irradiated human skin.
|
FOOTNOTES |
|---|
To whom correspondence should be addressed: University of Arizona, Arizona Cancer Center, 1515 N. Campbell Ave., Tucson, AZ 85724. Tel.: 520-626-5953; Fax: 520-626-8567; E-mail: elaine.jacobson{at}pharmacy.arizona.edu.
1 The abbreviations used are: ROS, reactive oxygen species; AGE, advanced glycation end product; BSA, bovine serum albumin; BSA-B6, pyridoxine-BSA conjugate; DABCO, 1,4-diazabicyclooctane; DES, desmosine; DHR 123, dihydrorhodamine 123; ECM, extracellular matrix; HBSS, Hanks' balanced salt solution; HP, hydroxypyridine; MALDI-TOF, matrix-assisted laser desorption time-of-flight; NE-3HP, N-ethyl-3-hydroxypyridinium cation; N-alkyl-3HP, N-alkyl-3-hydroxypyridinium; PN, pyridoxine; PI, propidium iodide; PYD, pyridinolin; RNAse A, ribonuclease A; SSL, solar simulated light; PBS, phosphate-buffered saline; MAP, mitogen-activated protein kinase.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
