Light-activated tissue bonding for excisional wound closure: a split-lesion clinical trial.

BACKGROUND: Apposition of wound edges by sutures provides a temporary scaffold and tension support for healing. We have developed a novel tissue-sealing technology, photoactivated tissue bonding (PTB), which immediately crosslinks proteins between tissue planes, thereby sealing on a molecular scale. OBJECTIVES: To determine the effectiveness of PTB for superficial closure of skin excisions and to compare the results with standard epidermal suturing. METHODS: A split-lesion, paired comparison study of 31 skin excisions was performed. Following deep closure with absorbable sutures, one-half of each wound was superficially closed with nonabsorbable nylon sutures while the other half was stained with Rose Bengal dye and treated with green light. Overall appearance and scar characteristics were rated at 2weeks and 6months in a blinded manner by three dermatologists viewing photographs, by two onsite physicians and by patients. RESULTS: At 2weeks, neither sutured nor PTB-treated segments showed dehiscence; however, PTB-sealed segments showed less erythema than sutured segments as determined by photographic (P=0·001) and onsite evaluations (P=0·005). Overall appearance after PTB was judged better than after sutures (P=0·002). At 6months, scars produced by PTB were deemed superior to scars resulting from sutures in terms of appearance (P<0·001), width (P=0·002) and healing (P=0·003). Patients were more satisfied with the appearance of the PTB-sealed wound half after 2weeks and 6months (P=0·013 and P=0·003, respectively). CONCLUSIONS: A novel molecular suturing technique produces effective wound sealing and less scarring than closure with nylon interrupted epidermal sutures. Comparisons with better suturing techniques are warranted.

Phototoxicity mechanism of a kryptocyanine dye in human red cell membranes and isolated murine mitochondria.

The phototoxicity mechanism of a kryptocyanine dye, N,N'-bis(2-ethyl-1,3-dioxolane)kryptocyanine (EDKC+), has been studied in RBC membranes and isolated mitochondria. Lipophilic, positively charged dyes, such as EDKC+, may be useful as tumor-cell-selective, light-activated cytotoxic agents. Exposure of the RBC membranes to 700-nm light and EDKC+ inhibited membrane acetylcholinesterase and photodecomposed EDKC+ in air-purged but not argon-purged samples. Photoinactivation of acetylcholinesterase was the same in D2O as in H2O and was not quenched by superoxide dismutase. Ascorbate and azide (10 mM) quenched or slightly enhanced, respectively, the inactivation. In argon-purged samples containing methyl viologen, EDKC+ photodecomposed, but acetylcholinesterase activity was unaffected. The mechanism may involve electron transfer to oxygen and subsequent formation of toxic photoproducts from EDKC+. In contrast, exposure of murine mitochondria to EDKC+ and 700-nm light caused inhibition of mitochondrial respiration in both the presence and absence of oxygen. The photodecomposition of EDKC+ correlated with inhibition of respiration. Thus, the phototoxicity of EDKC+ in mitochondria may be due to electron transfer from photoexcited EDKC+ to oxygen and electron acceptors in the membrane. These studies indicate that dyes such as EDKC+ may be useful for photochemotherapy of hypoxic regions in tumors.

Pyrimidine dimer formation and repair in human skin.

Cyclobutyl pyrimidine dimers have been detected in the DNA of human skin following in vivo irradiation with suberythemal doses of ultraviolet (UV) radiation from FS-20 sun lamp fluorescent tubes. Dimers were assayed by treatment of extracted DNA with Micrococcus luteus UV-specific endonuclease, alkaline agarose electrophoresis, and ethidium bromide staining. This technique, in contrast to conventional dimer assays, can be used with nonradioactive DNA and is optimal at low UV light doses. M. luteus endonuclease-sensitive sites were determined after exposure of untanned skin in two volunteers to UV light (0.97, 1.94, or 3.88 X 10(3) J/sq m; lambda, 290 to 360 nm). At 20 min postirradiation (dose, 1.94 X 10(3) J/sq m), fewer M. luteus endonuclease-sensitive sites were found in the DNA than immediately after the irradiation. Even fewer endonuclease-sensitive sites were found at 20 min when the UV-irradiated skin was subsequently irradiated with visible light than when the area was kept in the dark. These data suggest that some dimer disappearance by excision repair occurs within 20 min of UV irradiation and that photoreactivation of dimers can make a contribution to the total repair process.

DNA damage induced by 193-nm radiation in mammalian cells.

The contribution of DNA damage to the effects of 193-nm excimer laser radiation on mammalian cells in culture was studied in order to evaluate the mutagenic potential of this UV wavelength in vivo. Two approaches were taken: measurement of pyrimidine dimer-specific endonuclease-sensitive sites/megabase and comparison of the 193-nm radiation-induced cytotoxicity in normal versus DNA repair-deficient cells. The formation of pyrimidine dimer-specific endonuclease-sensitive sites/megabase was inversely related to the thickness of the cytoplasm overlying the nuclei of normal human fibroblasts (NHF) and Chinese hamster ovary cells. The results of these measurements and a calculation of the absorption coefficient of cytoplasm indicate that each 1 micron of cytoplasm attenuates the incident radiation by greater than 90% and, therefore, the nuclear DNA in tissue will be highly protected from 193-nm radiation by overlying cytoplasm. The reduction in colony-forming ability induced by 254-nm, 193-nm, and X-ray radiation was measured in NHF, xeroderma pigmentosum (group A) cells, and ataxia telangiectasia cells. Xeroderma pigmentosum (group A) cells were 16.5 times more sensitive to 254-nm radiation but only 3.5 times more sensitive to 193-nm radiation than NHF cells, indicating that cyclobutylpyrimidine dimers were not the major lethal lesion formed at 193 nm. AT cells were 3.4 times more sensitive to X-rays than NHF cells, but these cell types were almost equally sensitive to 193-nm radiation, indicating that 193 nm did not induce the same type of lethal lesions as X-rays.

Cytotoxicity and mutagenicity of low intensity, 248 and 193 nm excimer laser radiation in mammalian cells.

The cytotoxicity of 193 and 248 nm excimer laser radiation was compared to that produced by a germicidal lamp (predominantly 254 nm) using Chinese hamster ovary cells (CHO), and a human diploid fibroblast line, AG-1522A. Excimer laser radiation at 248 nm (3.5 X 10(2) w/m2) and germicidal radiation (5.3 X 10(-5) w/m2) caused toxicity in both cell lines, with the AG-1522A cells (D37 = 7-8 J/m2) being slightly more sensitive than the CHO cells (D37 = 11 J/m2). Incident 193 nm radiation was less cytotoxic than 248 nm to AG-1522A and CHO cells with D37 values of 18 and 85 J/m2, respectively. The mutagenic potential of UV excimer radiation at 193 and 248 nm was evaluated using the hypoxanthine guanine phosphoribosyl transfer assay system with CHO cells. Excimer laser radiation at 248 nm induced mutation in proportion to dose (1.7 X 10(-5) resistant colonies per survivor per J/m2 incident radiation) up to 14 J/m2, similar to results reported for 254 nm light. However, excimer laser radiation at 193 nm did not cause mutation greater than the dark control. The decreased cytotoxicity and mutagenicity of 193 nm radiation may be due to the shielding of the nucleus by cytoplasmic and membrane components or to the formation of different DNA photoproducts. These differences between 193 and 248 nm radiation may be important in choosing an excimer wavelength for ablation in biological systems.

Protein kinase C inhibits singlet oxygen-induced apoptosis by decreasing caspase-8 activation.

Although activation of protein kinase C (PKC) inhibits apoptosis induced by a variety of stimuli including singlet oxygen, the step at which PKC activation interferes with apoptotic signaling is not well defined. We have shown previously that caspase-8 and p38 mediate singlet oxygen-induced apoptosis in HL-60 cells. In this study, we investigated the influence of PKC on regulation of the caspase and p38 pathways initiated by singlet oxygen. Singlet oxygen induced Fas clustering and subsequent recruitment of FADD and caspase-8. Treatment of cells with the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA), a PKC activator, did not affect the binding of caspase-8 to the aggregated Fas. Surprisingly, under the same conditions PKC activation was still able to prevent singlet oxygen-induced activation of caspase-8 and block its downstream signaling events including cleavage of Bid and caspase-3, decrease in mitochondrial transmembrane potential and release of cytochrome c from mitochondria. Inhibition of PKC by GF109203 or H7 counteracted the TPA-mediated effects on the cleavage of caspases -3 and -8. However, neither activation nor inhibition of PKC affected p38 phosphorylation. These data indicate that PKC inhibits singlet oxygen-induced apoptosis by blocking activation of caspase-8.

Influence of dye and protein location on photosensitization of the plasma membrane.

Two membrane-photosensitizing dyes were used to investigate whether selected sites in the plasma membrane vary in their sensitivity to damage by singlet oxygen (1O2*) and, if so, what factors are responsible for the variation. The relative ability of Rose bengal (RB) and merocyanine 540 (MC540), both of which localize in the plasma membrane and produce 1O2*, to photosensitize five plasma membrane functions in P388D1 cells was evaluated. The five membrane functions assessed were: plasma membrane potential, proline transport, facilitated glucose diffusion, 5'-nucleotidase activity, and dye exclusion. Photosensitization efficiency by RB varied by a factor of 188 for these membrane functions, whereas for MC540 a range of only 24 was found. RB was a more efficient photosensitizer than MC540 but the relative efficiencies varied with the membrane function. The wide range of P50 values for RB suggests that it binds selectively to membrane sites where it causes damage with high efficiency; possibly a non-1O2* mechanism is involved. In contrast, MC540 photosensitized the three membrane functions involving integral membrane proteins about equally suggesting that differences are due to small variations in the distribution of MC540 in the plasma membrane and/or variations in the inherent reactivity of the membrane targets with 1O2*. The results indicate that the lability of membrane sites to photosensitization depends both on their inherent reactivity with 1O2* and the relative location of specific protein and dye molecules.

Comparison of photosensitized plasma membrane damage caused by singlet oxygen and free radicals.

The efficiency and selectivity of photosensitized damage to membrane functions may be influenced strongly by the identity of the initial reactive species formed by the photosensitizer. To test this possibility, a photosensitizer, rose bengal (RB), was used that resides in the plasma membrane and which generates singlet molecular oxygen (1O2*) upon excitation with visible light, and radicals plus 1O2* upon excitation with UV radiation. With this approach, 1O2* and radicals are formed at the same locations in the plasma membrane. The response of three plasma membrane functions, namely, proline transport, membrane potential, and membrane impermeability to charged dye molecules, was assessed. The efficiencies of the responses in the presence and absence of oxygen were compared per photon absorbed by RB at two wavelengths, 355 nm (UV excitation) and 532 nm (visible excitation). The efficiency of oxygen removal before irradiation was assessed by measuring the RB triplet lifetime. The three membrane functions were inhibited more efficiently at 355 nm than at 532 nm in the presence of oxygen indicating that the radicals are more effective at initiating damage to membrane components than 1O2*. The ratio of photosensitized effects at the two wavelengths in the presence of oxygen was the same for two membrane functions but not for the third suggesting that 1O2* and radicals initiate a common mechanistic pathway for damage to some membrane functions but not to others. Removing oxygen reduced the efficiency of 355 nm-induced photosensitization by factors of 1.4 to 7. The sensitivity of the three membrane functions to 1O2*-initiated damage varied over a factor of 50 whereas radical initiated damage only varied by a factor of 15. In summary, these results indicate that radicals and 1O2* formed at the same locations in the plasma membrane vary in their efficiency and specificity for membrane damage but may, in some cases, operate by a common secondary damage mechanism in the presence of oxygen.

Laser intensity and wavelength dependence of Rose-Bengal-photosensitized inhibition of red blood cell acetylcholinesterase.

The intensity and wavelength-dependence of Rose-Bengal-mediated photoinhibition of red blood cell acetylcholinesterase has been studied. Irradiation of dye-membrane suspensions with 308 nm laser excitation resulted in enzyme inhibition almost 50% greater than that obtained with 514 nm laser excitation. Sodium azide and argon purging greatly decreased the photosensitized enzyme inhibition at both wavelengths. Although Rose Bengal photosensitized enzyme inhibition more efficiently upon excitation into Sn (308 nm) than into S1 (514 nm), Stern-Volmer analysis of sodium azide quenching data gave similar quenching efficiencies at both wavelengths. Irradiation of dye-membrane suspensions with increasing intensities (Nd:YAG, 532 nm, 40 ps pulse duration) resulted in a decrease in enzyme inhibition. Saturation of the Rose Bengal fluorescence intensity and light transmission occurred with nearly the same intensity-dependence, suggesting that ground-state depletion occurs at the higher intensities. Our results demonstrate that excitation of a sensitizer into higher-lying excited singlet states can result in enhanced sensitizing efficiency. However, attempts to populate such states in Rose Bengal by sequential two-photon absorption using high intensities resulted only in ground-state depletion.

A wavelength dependent mechanism for rose bengal-sensitized photoinhibition of red cell acetylcholinesterase.

A 2-fold enhancement in the efficiency of rose bengal-photosensitized inhibition of red cell acetylcholinesterase activity was observed upon excitation of the dye in the ultraviolet (UV) (313 nm) compared to irradiation in the visible (514 or 550 nm). The measurements of efficiency of photosensitized enzyme inhibition were based on the effect produced when the same number of photons are absorbed by rose bengal (RB) at each wavelength. The mechanism for this unexpected enhancement of RB photosensitization upon UV excitation was investigated. The yield of singlet oxygen (O2(1 delta g], detected by time-resolved luminescence at 1270 nm, was independent of excitation wavelength for RB. Radicals were produced upon irradiation of RB at 313 nm but not at 514 nm as detected by bleaching of N,N-dimethylnitrosoaniline (RNO). Irradiation of RB at 313 nm but not at 514 nm appeared to cause homolytic cleavage of carbon-iodine bonds in the dye because iodine radicals, I, detected as I2 were produced with a quantum yield of 0.0041 +/- 0.0005 upon excitation in the UV. Photolysis of I2 in the presence of RNO caused bleaching of the RNO absorption at 440 nm, apparently resulting from reaction of I with RNO. Thus, the enhanced photosensitization upon UV excitation of RB is attributed to formation of I and/or RB. These results indicate that radicals, produced with low relative yield but having high reactivity compared to O2(1 delta g), can contribute to photosensitized enzyme inhibition and may represent an alternative mechanism for photodynamic therapy.

Apoptosis induced by dithiothreitol in HL-60 cells shows early activation of caspase 3 and is independent of mitochondria.

Previous studies have shown that under certain conditions some thiol-containing compounds can cause apoptosis in a number of different cell lines. Herein, we investigated the apoptotic pathways in HL-60 cells triggered by dithiothreitol (DTT), used as a model thiol compound, and tested the hypothesis that thiols cause apoptosis via production of hydrogen peroxide (H2O2) during thiol oxidation. The results show that, unlike H2O2, DTT does not induce apoptosis via a mitochondrial pathway. This is demonstrated by the absence of early cytochrome c release from mitochondria into the cytosol, the lack of mitochondrial membrane depolarization at early times, and the minor role of caspase 9 in DTT-induced apoptosis. The first caspase activity detectable in DTT-treated cells is caspase 3, which is increased significantly 1 - 2 h after the start of DTT treatment. This was shown by following the cleavage of both a natural substrate, DFF-45/ICAD, and a synthetic fluorescent substrate, z-DEVD-AFC. Cleavage of substrates of caspases 2 and 8, known as initiator caspases, does not start until 3 - 4 h after DTT exposure, well after caspase 3 has become active and at a time when apoptosis is in late stages, as shown by the occurrence of DNA fragmentation to oligonucleosomal-sized pieces. Although oxidizing DTT can produce H2O2, data presented here indicate that DTT-induced apoptosis is not mediated by production of H2O2 and occurs via a novel pathway that involves activation of caspase 3 at early stages, prior to activation of the common 'initiator' caspases 2, 8 and 9.

Action spectrum and mechanisms of UV radiation-induced injury in lupus erythematosus.

Photosensitivity associated with lupus erythematosus (LE) is well established. The photobiologic basis for this abnormal response to ultraviolet radiation, however, has not been determined. This paper summarizes the criteria for elucidating possible photobiologic mechanisms and reviews the literature relevant to the mechanism of photosensitivity in LE. In patients with LE, photosensitivity to wavelengths shorter than 320 nm has been demonstrated; wavelengths longer than 320 nm have not been adequately evaluated. DNA is a possible chromophore for photosensitivity below 320 nm. UV irradiation of skin produces thymine photodimers in DNA. UV-irradiated DNA is more antigenic than native DNA and the antigenicity of UV-irradiated DNA has been proposed, but not proven, to be involved in the development of clinical lesions. UV irradiation of mice previously injected with anti-UV-DNA antibodies produces Ig deposition and complement fixation that appears to be similar to the changes seen in lupus lesions. Antibodies to UV-irradiated DNA occur in the serum of LE patients although a correlation between antibody titers and photosensitivity was not observed. Defective repair of UV-induced DNA damage does not appear to be a mechanism for the photosensitivity in LE. Other mechanisms must also be considered. The chromophore for photosensitivity induced by wavelengths longer than 320 nm has not been investigated in vivo. In vitro studies indicate that 360-400 nm radiation activates a photosensitizing compound in the lymphocytes and serum of LE patients and causes chromosomal aberrations and cell death. The mechanism appears to involve superoxide anion. Further research is required to establish the action spectrum for long wavelength photosensitivity in vivo and to elucidate the mechanisms for the photosensitivity at all wavelengths.

Photoreactions of 3,3',4',5-tetrachlorosalicylanilide with proteins.

The photoreactions of 3,3',4',5-tetrachlorosalicylanilide anion (TCSA-) with two serum proteins were studied. TCSA- and anions of two similar compounds, N-ethyl-3,5-dichlorosalicylamide and salicylanilide, bind noncovalently to human serum albumin (HSA) without irradiation in aqueous pH 7.4 buffered solutions. TCSA- noncovalently bound to HSA yields three types of photoproducts when irradiated with ultraviolet light (lambda greater than 360 nm). A covalently bonded photoadduct between TCSA- and HSA is formed and histidines in HSA are chemically modified. In addition to these two types of photoproducts which involve HSA, two of the four TCSA- photoproducts which form when HSA is absent are also formed when the TCSA-/HSA complex is irradiated. The results presented indicate that not all proteins in the skin are capable of being the carrier protein in photoallergy of TCSA- and that cross reactivity to other halogenated salicylanilides can be explained by further photochemical reactions of TCSA- photoproducts.

Experimental photoaging in C3H/HeN, C3H/HeJ, and Balb/c mice: comparison of changes in extracellular matrix components and mast cell numbers.

Chronic exposure of human or murine skin to ultraviolet B (UVB) radiation alters dermal extracellular matrix composition and increases the number of mast cells and inflammatory cells. Experiments were designed to test the possible role of UVB-induced tumor necrosis factor-alpha in these photoaging changes based on reports that C3H/HeN, but not C3H/HeJ or Balb/c mice, produce excess TNF-alpha in response to UVB exposure. Pigmented C3H/HeN and C3H/HeJ strains were exposed to a total of 75 J/cm2 of UVB radiation, and unpigmented Balb/c mice were exposed to 19 J/cm2. The UVB-induced increases in collagen, glycosaminoglycans, and neutrophil number were similar or the same in all three strains. The elastin increase was greater in C3H/HeJ than in C3H/HeN mice. The most striking difference between the strains was a 7.7-fold UVB-induced increase in mast cells in C3H/HeN mice compared to no increase in irradiated C3H/HeJ mice and a 2.3-fold increase in Balb/c mice. These results suggest that excess TNF-alpha (or other mediator) produced in C3H/HeN skin (but not C3H/HeJ skin) in response to UVB exposure is involved in the mast cell increase and partial inhibition of elastin increase, but that neither these mediators nor mast cell products are important mediators for the chronic UVB-induced increases in neutrophils, glycosaminoglycans, and collagen. When a possible source of the excess TNF-alpha was investigated, it was found that isolated epidermal cells from all three strains produced increases in TNF-alpha in response to UVB radiation. These results, as well as the previous results showing differences between these strains in UVB-induced effects on cutaneous immune function, are consistent with a model in which UVB-induced mediators from the epidermis stimulate another cell type to produce excess TNF-alpha (and other mediators) in the C3H/HeN but not C3H/HeJ or Balb/c mice.

Benoxaprofen photosensitization of cell membrane disruption.

Benoxaprofen (BXP) is a nonsteroidal anti-inflammatory drug which causes cutaneous phototoxicity. Because the in vivo phototoxicity may involve photosensitized damage to mast cell membranes, the mechanism for photosensitized damage was studied in a single model system, the red blood cell. Oxygen-dependent and oxygen-independent mechanisms for membrane disruption were detected. Oxygen-dependent lysis was not quenched by superoxide dismutase and was quenched only at high sodium azide concentrations. A two-step mechanism is proposed involving initial photodecarboxylation of BXP to form a lipophilic photoproduct which subsequently photosensitizes membrane damage. Human serum albumin at 0.03% totally inhibited BXP-photosensitized lysis.

Unscheduled DNA synthesis in human skin after in vitro ultraviolet-excimer laser ablation.

DNA damage repaired by the excision repair system and measured as unscheduled DNA synthesis (UDS) was assessed in freshly excised human skin after 193 and 248 nm ultraviolet (UV)-excimer laser ablative incisions. Laser irradiation at 248 nm induced DNA damage throughout a zone of cells surrounding the ablated and heat-damaged area. In contrast, with 193 nm irradiation UDS was not detected in cells adjacent to the ablated area, even though DNA strongly absorbs this wavelength. Our results suggest that the lack of UDS after 193 nm irradiation is due to: "shielding" of DNA by the cellular interstitium, membrane, and cytoplasm, DNA damage that is not repaired by excision repair, or thermal effects that either temporarily or permanently inhibit the excision repair processes.

Effects of systemic indomethacin, meclizine, and BW755C on chronic ultraviolet B-induced effects in hairless mouse skin.

Chronic exposure of hairless mice to ultraviolet B (UVB) radiation is associated with inflammation as well as an altered macromolecular composition of the dermis. This study was designed to determine whether or not various systemic anti-inflammatory agents inhibit chronic UVB-induced changes in the macromolecular content of the dermis and, if so, whether each agent had the same or different effects. The agents and doses were chosen for their ability to inhibit the changes induced by a single exposure to UVB radiation (increased vasopermeability, neutrophil accumulation, and skin-fold thickness). Indomethacin, a cyclooxygenase inhibitor, and meclizine, an H1 histamine receptor antagonist, were administered from slow-release pellets. BW755C, a combined cyclooxygenase and lipoxygenase inhibitor, was administered intraperitoneally 30 min prior to UVB exposure. Animals were exposed to UVB three times per week for 20-26 weeks or were unirradiated. The elastin, glycosaminoglycan and collagen content of the skin were determined by measuring the desmosine, uronic acid, and hydroxyproline levels, respectively. The amount of each macromolecule per area of skin increased after chronic UVB exposure. The increase in desmosine was inhibited by indomethacin; the increase in hydroxyproline was inhibited by meclizine and BW755C. None of the agents inhibited the uronic acid increase. These results suggest that chronic inflammation contributes to the dermal changes seen in chronically UVB-exposed skin and that different inflammatory mediators are involved in the increases observed in elastin, glycosaminoglycans, and collagen.

The effects of radicals compared with UVB as initiating species for the induction of chronic cutaneous photodamage.

There is substantial evidence that ultraviolet radiation induces the formation of reactive oxygen species which are implicated as toxic intermediates in the pathogenesis of photoaging. The aim of this study was to determine whether repeated topical treatment with benzoyl peroxide, a source of free radicals, produced the same cutaneous effects as chronic ultraviolet B radiation. Three concentrations of benzoyl peroxide (0.1, 1.5, 5.0% wt/wt) and three cumulative fluences of ultraviolet B radiation (0.9, 2.2, 5.1 J per cm2) used alone and in all combinations along with appropriate controls. Female SKH1 (hr/hr) albino hairless mice were treated 5 d per wk for 12 wk. Extracellular matrix molecules and histologic parameters were assessed. Ultraviolet B radiation induced a fluence-dependent and time-dependent increase in skin-fold thickness. Fluence dependence was seen for epidermal thickness, sunburn cell numbers, dermal thickness, glycosaminoglycan content, mast cell numbers, and skin-fold thickness. Benzoyl peroxide treatment alone caused less marked increases in epidermal and dermal measures compared with ultraviolet B under the conditions used. A benzoyl peroxide concentration-dependent increase was only observed for elastin content, although the highest concentration of benzoyl peroxide increased epidermal thickness and glycosaminoglycan content. A synergistic interaction between ultraviolet B and benzoyl peroxide was not found. These results indicate that repeated administration of benzoyl peroxide produces skin changes in the hairless mouse that qualitatively resemble those produced by ultraviolet B and suggest that common mechanisms may be involved. In addition, any potential synergistic effect of ultraviolet B and benzoyl peroxide was below the level of detection used in this study.

Mechanism of tetracycline phototoxicity.

Studies were made to determine factors important in the phototoxicity mechanism of 7 clinically used tetracyclines (TC). The clinical phototoxicity, the rates of photochemical degradation, and the in vitro phototoxicity of the TCs were qualitatively but not quantitatively correlated. Phototoxicity in vitro was partially oxygen-dependent and possibly singlet oxygen is involved. The contribution of photoproducts to the phototoxic process may be the basis for the reported differences between the in vivo action spectrum and the absorption spectrum of demethylchlorotetracycline. A mechanistic model for in vivo phototoxicity is proposed where the absorption of UVA radiation by TC leads to at least two main processes: (i) photosensitization by the drug of biologic molecules to cause phototoxicity; (ii) production of one or more photoproducts which photosensitize by absorption of visible radiation.

Endogenous skin fluorescence includes bands that may serve as quantitative markers of aging and photoaging.

Aging and photoaging cause distinct changes in skin cells and extracellular matrix. Changes in hairless mouse skin as a function of age and chronic UVB exposure were investigated by fluorescence excitation spectroscopy. Fluorescence excitation spectra were measured in vivo, on heat-separated epidermis and dermis, and on extracts of mouse skin to characterize the absorption spectra of the emitting chromophores. Fluorescence excitation spectra obtained in vivo on 6 wk old mouse skin had maxima at 295, 340, and 360 nm; the 295 nm band was the dominant band. Using heat separated tissue, the 295 nm band predominantly originated in the epidermis and the bands at 340 and 360 nm originated in the dermis. The 295 nm band was assigned to tryptophan fluorescence, the 340 nm band to pepsin digestable collagen cross-links fluorescence and the 360 nm band to collagenase digestable collagen cross-links fluorescence. Fluorescence excitation maxima remained unchanged in chronologically aged mice (34-38 wk old), whereas the 295 nm band decreased in intensity with age and the 340 nm band increased in intensity with age. In contrast, fluorescence excitation spectra of chronically UVB exposed mice showed a large increase in the 295 nm band compared with age-matched controls and the bands at 340 and 350 nm were no longer distinct. Two new bands appeared in the chronically exposed mice at 270 nm and at 305 nm. These reproducible changes in skin autofluorescence suggest that aging causes predictable alterations in both epidermal and dermal fluorescence, whereas chronic UV exposure induces the appearance of new fluorphores.