Artigo Acesso aberto Revisado por pares

Non-Coherent Near Infrared Radiation Protects Normal Human Dermal Fibroblasts from Solar Ultraviolet Toxicity

1998; Elsevier BV; Volume: 111; Issue: 4 Linguagem: Inglês

10.1046/j.1523-1747.1998.00338.x

ISSN

1523-1747

Autores

Salatiel Menezes, Bernard Coulomb, Corinne Lebreton, Louis Dubertret,

Tópico(s)

Spaceflight effects on biology

Resumo

The sun is the most important and universal source of non-ionizing radiation shed on human populations. Life evolved on Earth bathed by this radiation. Solar UV damages cells, leading to deleterious conditions such as photoaging and carcinogenesis in human skin. During the process of evolution, the cells selected dark- and light-dependent repair mechanisms as a defence against these hazardous effects. This study describes the induction by non-coherent infrared radiation (700–2000 nm), in the absence of rising temperature, of a strong cellular defense against solar UV cytotoxicity as well as induction of cell mitosis. Blocking mitoses with arabinoside-cytosine or protein synthesis with cycloheximide did not abolish the protection, leading to the conclusion that this protection is independent of cell division and of protein neosynthesis. The protection provided by infrared radiation against solar UV radiation is shown to be a long-lasting (at least 24 h) and cumulatif phenomenon. Infrared radiation does not protect the lipids in cellular membranes against UVA induced peroxidation. The protection is not mediated by heat shock proteins. Living organisms on the Earth's surface are bathed by infrared radiation every day, before being submitted to solar UV. Thus, we propose that this as yet undescribed natural process of cell protection against solar UV, acquired and preserved through evolutional selection, plays an important role in life maintenance. Understanding and controlling this mechanism could provide important keys to the prevention of solar UV damage of human skin. The sun is the most important and universal source of non-ionizing radiation shed on human populations. Life evolved on Earth bathed by this radiation. Solar UV damages cells, leading to deleterious conditions such as photoaging and carcinogenesis in human skin. During the process of evolution, the cells selected dark- and light-dependent repair mechanisms as a defence against these hazardous effects. This study describes the induction by non-coherent infrared radiation (700–2000 nm), in the absence of rising temperature, of a strong cellular defense against solar UV cytotoxicity as well as induction of cell mitosis. Blocking mitoses with arabinoside-cytosine or protein synthesis with cycloheximide did not abolish the protection, leading to the conclusion that this protection is independent of cell division and of protein neosynthesis. The protection provided by infrared radiation against solar UV radiation is shown to be a long-lasting (at least 24 h) and cumulatif phenomenon. Infrared radiation does not protect the lipids in cellular membranes against UVA induced peroxidation. The protection is not mediated by heat shock proteins. Living organisms on the Earth's surface are bathed by infrared radiation every day, before being submitted to solar UV. Thus, we propose that this as yet undescribed natural process of cell protection against solar UV, acquired and preserved through evolutional selection, plays an important role in life maintenance. Understanding and controlling this mechanism could provide important keys to the prevention of solar UV damage of human skin. arabinosidecytosine infrared radiation heat shock protein thiobarbituric acid reacting substances The solar radiation that reaches the Earth's surface is composed of wavelengths ranging from 290 to 4000 nm, a scope that includes ultraviolet radiation (UVB, 290–320 nm; UVA, 320–400 nm), visible light (VL, 400–700 nm), and infrared radiation (IR, 700–4000 nm). Of this energy, 40% lies in the 700–4000 nm range, i.e., infrared radiation (Koller, 1965Koller L.R. Ultraviolet Radiation. John Wiley, New York1965: 105-152Google Scholar). Throughout evolution, living cells have been exposed to this polychromatic electromagnetic non-ionizing radiation. Some of these wavelengths are hazardous to the cells, damaging essential molecules such as nucleic acids and proteins. It is commonly accepted that solar UV is responsible for almost all deleterious photoinduced effects on human skin. UVA and UVB are believed to constitute major risk factors in the etiology of human skin cancers and photoaging (Gange et al., 1986Gange R.W. Park I.K. Accelleta M. Kagectsu N. Blakett A.D. Parrish J.A. Action spectra for cutaneous responses to ultraviolet radiations.in: Urbach F. George F.W. The Biological Effects of UV Radiation. Praegger Publishers, New York1986: 57-67Google Scholar;Kligman and Kligman, 1989Kligman L.H. Kligman A.M. The nature of photoaging: its prevention and repair.Photodermatol. 1989; 3: 215-217Google Scholar;Ananthaswami and Pierceall, 1990Ananthaswami H.N. Pierceall W.E. Molecular mechanisms of ultraviolet radiation.Photochem Photobiol. 1990; 52: 1119-1136Crossref Scopus (322) Google Scholar). At similar doses, UVA is far less mutagenic than is UVB (Drobetski et al., 1995Drobetski E.A. Turcotte J. Chateauneuf A. A role for ultraviolet A in solar mutagenesis.Proc Natl Acad Sci USA. 1995; 92: 2350-2354Crossref Scopus (225) Google Scholar), leading to the idea that UVA plays a minor role in deleterious solar UV-induced effects on human skin; however, UVA constitutes more than 90% of the solar UV energy reaching the Earth's surface and penetrates the basal layers of the epidermis and the dermis with great facility, thus compensating for its low dose-effect efficiency (Bruls et al., 1984Bruls W.A.G. Slaper H. van der Leun J.C. Berrens L. Transmission of human epidermis and stratum comeum as a function of thickness in the ultraviolet and visible wavelengths.Photochem Photobiol. 1984; 40: 485-494Crossref PubMed Scopus (390) Google Scholar). UVA has been shown to be mutagenic in cultured cells (Tyrrell and Keyse, 1990Tyrrell R.M. Keyse S.M. The interaction of UVA radiation with cultured cells.J Photochem Photobiol B Biol. 1990; 4: 349-361Crossref PubMed Scopus (258) Google Scholar), to induce malignant melanomas in melanin-pigmented blackfish more efficiently than UVB, when the criterion of comparison is the product of monochromatic efficiency and the sunlight fluence rate (Setlow, 1996Setlow R.B. Relevance of in vivo models in melanoma skin cancer.Photochem Photobiol. 1996; 63: 410-412Crossref PubMed Scopus (4) Google Scholar), and to induce skin cancers in hairless albino mice (De Gruiil et al., 1993De Gruiil F.R. Steremborg H.J.C.M. Forbes P.D. et al.Wavelength dependence of skin cancer induction by ultraviolet irradiation of albino hairless mice.Cancer Res. 1993; 53: 53-60Google Scholar). More significantly, a causal relationship between artificial UVA baths for cosmetic or medical purposes and a substantial increase in human melanomas has recently been demonstrated epidemiologically (Autier et al., 1994Autier P. Doré J.F. Lejeune F. et al.Cutaneous melanoma and exposure to sunlamps or sunbeds: an EORTC multicenter case-control study in Belgium, France and Germany.Int J Cancer. 1994; 58: 809-813Crossref PubMed Scopus (129) Google Scholar;Westerdahl et al., 1994Westerdahl J. Olsson H. Masbäck A. et al.Use of sunbeds and sunlamps and malignant melanoma in Southern Sweden.Am J Epidemiol. 1994; 140: 691-699PubMed Google Scholar). Moreover, UVA is able to produce singlet oxygen in the cellular environment (Basu-modak and Tyrrell, 1993Basu-modak S. Tyrrell R.M. Singlet oxigen: a primary effector in the ultraviolet A/near visible light induction of the human heme oxygenase gene.Cancer Res. 1993; 53: 4505-4510PubMed Google Scholar), which could explain its biologic effects and its greater efficiency as a mutagenic rather than as a cytotoxic agent (Robert et al., 1996Robert C. Muel B. Benoit A. Dubertret L. Sarasin A. Stary A. Cell survival and shuttle vector mutagenesis induced by ultraviolet A and ultraviolet B radiation in a human cell line.J Inv Dermatol. 1996; 106: 721-728Crossref PubMed Scopus (103) Google Scholar). Much of our knowledge about solar radiation effects comes from experiments with monochromatic UV. Consequently, the additive, synergistic, or antagonistic interactions between the different solar wavelengths have been largely overlooked. This is true particularly of the biologic effects of IR radiations, which have been neglected due to the idea that these effects are almost always mediated by heat enhancement, leading to the induction of heat shock proteins (HSP) (Lindquist, 1986Lindquist S. The heat shock response.Ann Rev Biochem. 1986; 55: 1151-1191Crossref PubMed Google Scholar;Maytin et al., 1993Maytin E.V. Murphy L.A. Merrill M.A. Hyperthermia induces resistance to ultraviolet light B in primary and immortalized murine keratinocytes.Cancer Res. 1993; 53: 4952-4959PubMed Google Scholar). The incidence of skin cancer detected in human populations submitted to high levels of solar radiation seems to be less than what would be expected taking into account the amount of damage inflicted on cellular DNA by solar UV fluence and the repair capability of the cells (Sutherland, 1996Sutherland B.M. Mutagenic lesions in photocarcinogenesis: induction and repair of pyrimidine dimers.Photochem Photobiol. 1996; 63: 375-377Crossref Scopus (11) Google Scholar). This means that (i) the solar UV damaging potential has been overestimated, (ii) the repair potential of the cells has been underestimated, or (iii) the effects of UV in a polychromatic light beam are not the same as those of monochromatic UV, due to as yet unknown antagonistic effects. Although sunlight is polychromatic, its final effect on human skin is the result of not only the action of each wavelength individually, but also the interactions between these wavelengths. In a previous study, we observed that E. coli bacteria pre-irradiated with non-lethal doses of non-coherent IR, in conditions preventing temperature augmentation, became more resistant to a posterior challenge by UVC (Menezes, unpublished data). In this study, in order to mimic the schedule of natural solar radiation exposure, we studied the effects of pre-irradiation with non-coherent near-IR at non-cytotoxic doses and under temperature-controlled conditions (25°C), on the cytotoxic effects of UVA and UVB on human skin cultured fibroblasts. Primary-culture fibroblasts were obtained from healthy, plastic breast surgery donors as described earlier (Le Panse et al., 1996Le Panse R. Bouchard B. Lebreton C. Coulomb B. Modulation of keratynocyte growth factor (KGF) mRNA expression in human dermal fibroblasts grown in monolayer or within a collagen matrix.Exp Dermatol. 1996; 5: 108-114Crossref PubMed Scopus (17) Google Scholar). Briefly, skin was cleaned of excess deep dermis and subcutaneous fat, cut into thin pieces, rinsed in Puck's-ethylenediamine tetraacetic acid balanced salt solution and incubated in Eagle's minimal essential medium supplemented with 10% fetal calf serum, 100 IU penicillin per ml, 100 μg streptomycin per ml, and 2.5 μg amphotericin B per ml. The fibroblasts were then propagated in EMEM with 10% fetal calf serum, without phenol red, antibiotics, or fungizone. For the experiments, cells between the fourth and eighth passages were seeded at densities of 2 × 105 cells per Petri dish (35 mm). All cultures were kept at 37°C in a 5%-CO2 humidified atmosphere. IR was obtained from an IR GE 27, 250 W lamp (at a distance of 42 cm), with an emission spectrum in the range of 400–2000 nm with very low UV emission, as described by the manufacturer. For IR irradiations, the Petri dishes with the cells in their culture medium (4 mm thickness) were placed on a water-cooled plate that maintained the temperature of the medium in the dishes between 20 and 25°C, as measured with a precision thermometer (Quick Novo, Germany, precision 0.2°C, coupled to a steel immersion probe). Two 4 mm thick glass windows (one polished and one frosted) and a long-wave pass sharp-cut filter (Schott RG 715) were interposed between the lamp and the Petri dishes to ensure homogeneous irradiation and to remove wavelengths shorter than 700 nm. No focusing lens was used, to avoid heat concentration. Under these conditions, 30 min of IR irradiation was not toxic to the cells, but displayed a maximum of photoprotection. The irradiance of the IR lamp, at the cells plane, was 45 mW per cm2 as measured with a thermopile (Müller GmbH Elektronik-Optic, Germany), which has a constant sensibility in the range 270–4000 nm. After 30 min, the fluence delivered to the cells was 810 kJ per m2. Prior to UV irradiations, cells grown to near confluence were washed twice with 2 ml of Hank's balanced salt solution (HBSS) for 5 and 15 min, respectively. UVA (365 nm) was obtained from an UVA illuminating table (TFP35.L, Vilbert-Lourmat, Marne-la-Vallée, France) containing six UVA lamps of 15 W. The spectral output of the UVA set (through the two 4 mm glass windows and the lid of the Petri dish) was in the range 325–400 nm (λmax, 365 nm), as described elsewhere (Tirache and Morlière, 1995Tirache I. Morlière P. Hydrogen peroxide and catalase in UVA-induced lipid peroxidation in cultured fibroblasts.Redox Report. 1995; 1: 105-111Google Scholar). The irradiance of the UVA table, at the level of the cells, was 45 W per m2. UVB (312 nm) was obtained from an UVB illuminating table (TFX-35.M, Vilbert-Lourmat) equipped with six 15 W UVB lamps. Due to its built-in filter, the emission spectrum of this UVB table is in the range 290–320 nm, with no UVC and a λmax at 312 nm. The fluence rate was 12.5 W per m2 at the level of the cells. Due to the short time of irradiation with the UVB Table (2 s), the low UVA contamination is irrelevant from a biologic point of view. UV irradiations were carried out at 37°C. Before UV irradiations, the culture medium of the cells was removed, identified to each dish and kept in sterile conditions to be reused after irradiations, thus avoiding growth stimuli by new medium or new serum. UV irradiations immediately followed IR exposures except when specified. UVA irradiations were conducted through the glass windows (to remove the few UVB emitted by the lamps), the lids of the dishes, and 1 ml HBSS (2 mm thickness). For UVB, the lids and the glass windows were removed and the cells covered with a thin layer of Hank's solution. UV fluences were determined by chemical actinometry, performed in the Petri dishes and based on the photoreduction of ferrioxalate. Before and after each experiment, UV fluences were checked over with UV radiometers VLX-365 and VLX-312 (Vilbert Lourmat). Control cells consisted of cells treated under the same conditions as irradiated cells, but kept in the dark (sham irradiation). Cell survival was measured by counting the viable cells (estimated with the Trypan Blue Test) 24 h (UVA) or 72 h (UVB) after treatments. This simple method has the advantage of taking into account the final cytotoxic effect of UV, whatever the mechanism of toxicity involved may be. To block mitosis, as required in some experiments, arabinoside-cytosine (Ara-C) at a final concentration of 1 μg per ml was added to the cellular medium after the different treatments and incubated for 24 h, before cell counting. This concentration blocked cell division completely, but was not cytotoxic (Coulomb et al., 1984Coulomb B. Dubertret L. Bell E. Touraine R. The contractility of fibroblasts in a collagen lattice is reduced by corticosteroids.J Invest Dermatol. 1984; 82: 341-344Abstract Full Text PDF PubMed Scopus (39) Google Scholar). The results are expressed as the mean of three independent experiments ± SD. Twenty-four hours after irradiations, 5 μg per ml (final concentration) of bisbenzidine (Hoechst 33.342) was added to each Petri dish and incubated for 1 h at 37°C. The cells were then examined with an inverted microscope in fluorescence mode for visualization of the nuclei, or in phase contrast for visualization of the mitotic cells. Each point of irradiation consisted of three dishes, and three random counts per dish were performed. The total number of cells and the number of mitotic cells were determined, with the results presented as the mean of three independent experiments ± SD. Iminediately after irradiations, cycloheximide (Sigma, Saint Quentin Fallavier, France) diluted in distilled water was added to the culture medium at a final concentration of 10 μg per ml and incubated for 24 h, before cell counts. In this concentration, cycloheximide blocked protein synthesis and induced an approximate 30% decrease in the number of cells. The production of TBARS was evaluated as described byMorlière et al., 1991Morlière P. Moysan A. Sanctus R. Hupe G. Mazière J.C. Dubertret L. UVA-induced lipid peroxidation in cultured human skin fibroblasts.Biochim Biophys Acta. 1991; 1084: 261-268Crossref PubMed Scopus (139) Google Scholar. Briefly, after irradiations, 900 μl aliquots of the supernatants were collected, added to 100 μl of butylated hydroxytoluene (2% vol/vol in ethanol), and heated to 80°C in the presence of thiobarbituric acid. The TBARS were extracted with 1-butanol and quantitated fluorimetrically by excitation with 515 nm and emission recording at 550 nm in a Spex 112 spectrofluorometer (Jobin-Yvon, Longjumeau, France). Determinations of TBARS were performed in triplicate and the results given as the mean of three independent experiments ± SD. Cells were seeded in Labtek chamber slides (eight wells) at densities of 1.6 × 104 cells per well, and cultured for 3 d under the same conditions as cultures in Petri dishes. The cells were then washed twice with HBSS, irradiated as required, and fixed in 100% methanol at –20°C. HSP induction was evaluated using a mouse monoclonal antibody specific for 72 kDa HSP protein (Amersham, Les Ulis, France) and fluorescent isothiocyanate-labeled goat anti-mouse IgG (Biosys, Compiègne, France) as a second antibody. The slides were then examined and photographed in a fluorescence microscope. In preliminary experiments we determined that, under our experimental conditions, 30 min irradiation with the IR lamp (810 kJ per m2) elicited the maximal protective response, without any measurable cytotoxicity. Irradiation with 250 kJ UVA per m2 decreased the number of viable cells of 45% as compared with non-irradiated control cells. When the cells were pre-irradiated with IR, the same dose of UVA decreased this number by only 15% (Figure 1). Similarly, 500 J UVB per m2 decreased the number of viable cells by 75%, whereas the number of cells pre-irradiated with IR decreased by only 45% (Figure 1). Time-course experiments show that the protection induced by IR is detectable almost immediately after IR irradiation, is enhanced progressively until reaching a maximum 24 h later, and then decreases, disappearing almost completely within 3 d This temporal profile of the protection induced by IR is shown in Figure 2 for UVA The same pattern was found for UVB (data not shown) Fibroblasts were irradiated one, two, or three times with IR, before irradiation with UVA Between the IR irradiations the cells were incubated for 3 h in standard conditions, with their original medium Figure 3 shows that the proliferating stimulus of one, two, or three irradiations with IR was very similar; 250 kJ UVA per m2 led to a decrease of about 50% in the number of cells It can be seen in this figure that one pre-irradiation with IR inhibited the cytotoxic effect of UVA, leading to a decrease of only 20% in the number of cells Two IR irradiations abolished the effect of UVA almost completely, leading to a loss of only 2% of the cells, and three IR irradiations abolished the cytotoxicity of this dose of UVA completely, indicating that the protection provided by IR is cumulative We found the same cumulative protection against UVB (data not shown) A single experiment was carried out to calculate the dose reduction factor after three IR irradiations and subsequent variable doses of UVA or UVB For a decrease of 50% in the number of viable cells, we found dose reduction factors of 215 (UVA) and 26 (UVB); however, these figures concern cells from a single donor, which displayed strong protection Thus, they do not take into account the intrinsic differences of cells from different donors In our experiments IR irradiations systematically led to an augmentation of the number of cells, as compared with the non-irradiated, control cells. On microscopic examination, many more mitotic cells were seen in IR irradiated cultures, even when these cultures were in confluence or were UV irradiated, situations not favoring cell division. Thus, an experiment was designed to quantitate mitosis induction by IR radiation. Table 1 shows that the cultures irradiated with IR have 2-fold as many cells in mitosis as control cultures. UVA alone inhibited mitosis completely, whereas IR antagonised this UVA effect. This increase in the number of mitotic cells could be visualized as early as 4 h after IR irradiation, reaching a maximum 24 h later.Table 1Near-IR induces mitosis of human skin fibroblastsPercentage of mitotic cellsaValues represent the mean ± SD of three independent experiments in triplicate.Control3.06 ± 0.58IR (30 min)6.39 ± 1.15UVA (250 kJ per m2)0.00 ± 1.15IR + UVA2.32 ± 0.58a Values represent the mean ± SD of three independent experiments in triplicate. Open table in a new tab To determine if the protection induced by IR was linked only to the induction of mitosis, we performed experiments in which the mitotic cycle was inhibited by Ara-C. Figure 4 shows that Ara-C at 1 μg per ml was not cytotoxic, but prevented division of the cells irradiated or not with IR, for the same extent. Furthermore, this figure shows that Ara-C did not modify the protection induced by IR against the cytotoxicity of UVA, in the conditions of this experiment. To investigate whether the IR-induced protection is constitutive or is dependent on the neosynthesis of proteins, we blocked protein synthesis with cycloheximide. Figure 5 shows that cycloheximide induced a decrease of about 30% in the number of non-irradiated cells, 20% in the number of IR irradiated cells, and 15% in the number of UVA irradiated cells, but was not able to decrease the IR-induced protection. One of the best described effects of UVA on human skin fibroblasts is the peroxidation of lipids in the cellular membrane (Morlière et al., 1991Morlière P. Moysan A. Sanctus R. Hupe G. Mazière J.C. Dubertret L. UVA-induced lipid peroxidation in cultured human skin fibroblasts.Biochim Biophys Acta. 1991; 1084: 261-268Crossref PubMed Scopus (139) Google Scholar). Even if this lipid peroxidation is unlikely to be responsible for UVA cytotoxicity, we decided to investigate whether IR radiations could protect the cells from lipid peroxidations. Our results show clearly that preirradiation with IR does not change the amount of TBARS produced by UVA (Table 2).Table 2IR does not protect membrane lipids from UVA-induced peroxidationTBARS (pM per cell)aValues represent the mean ± SD of three independent experiments in triplicate.Control1.0 × 10–5 ± 0.22 × 10–5IR (30 min)0.9 × 10–5 ± 0.1 × 10–5UVA (250 kJ per m2)41.9 × 10–5 ± 1.6 × 10–5IR + UVA35.6 × 10–5 ± 0.97 × 10–5a Values represent the mean ± SD of three independent experiments in triplicate. Open table in a new tab Living cells evolved in an environment bathed by solar polychromatic radiation, where some wavelengths are strongly absorbed by crucial molecules such as nucleic acid and proteins. This absorption triggers photochemical reactions that damage these molecules, thus challenging both survival and maintenance of hereditary information. The cells developed adaptive responses and enzymatic repair systems to assure these vital functions, reaching an equilibrium between damage and repair in normal environmental conditions. Many of these repair systems have been described and their enzymatic mechanisms are well understood; however, cellular responses to solar radiations are more complex than it was supposed, due to interactions between different wavelengths. In this study we demonstrate that pre-irradiation with non-coherent near-IR at non-cytotoxic doses and in the absence of temperature raising protects human skin fibroblasts from the cytotoxic effects of UVA and UVB (Figure 1). Interposition of long wave pass cut filter RG 715, which cuts off wavelengths shorter than 700 nm, between the cells and the lamp did not change the protective effect. As the lamp does not emit wavelengths longer than 2000 nm, we conclude that the protection is triggered by radiation in the range 700–2000 nm, i.e., near-IR. Actually, the emission of the lamp in the range 1000–2000 nm is very weak, suggesting that the protection is most probably provided by wavelengths between 700 and 1000 nm. Figure 2 shows a Gaussian time course profile of the protection, indicating an inducible process. The immediate appearance of the protection and its cumulative characteristic (Figure 3), however, could mean that this is a constitutive, adaptive response, triggered somehow by the photonic energy of near-IR. Other similarly dual (constitutive-induced) systems of cellular protection have been previously described, e.g., the SOS repair system in E. coli bacteria (Walker, 1995Walker G.C. SOS-regulated proteins in translesion DNA synthesis and mutagenesis.Tibs. 1995; 20: 416-420Abstract Full Text PDF PubMed Scopus (88) Google Scholar). The results shown in Figure 3 and Table 1 are in agreement with those ofContinenza et al., 1993Continenza M. Ricciardi G. Franchitto A. Effects of low power 904 nm radiation on rat fibroblasts explanted and cultured in vitro.J Photochem Photobiol B Biol. 1993; 19: 231-234Crossref Scopus (5) Google Scholar. These authors showed that irradiation of rat fibroblasts in culture with 904 nm low power laser radiation improved the growth of the cultures. The best stimulation was obtained when they used repeated irradiations, whatever the duration of each irradiation was. Our data do not permit to know by which mechanism IR radiation induces cell division. It has been shown that irradiation of human peripheral lymphocytes with 820 nm laser light increases the level of ATP in these cells (Herbert et al., 1989Herbert K.E. Bhusatte L.L. Scott D.L. Diarnantopoulos C. Perret D. Effect of laser light at 820 nm on adenosine nucleotide levels in human lymphocytes.Lasers Life Sci. 1989; 3: 37-46Google Scholar). In mammalian cells, the division cycle is dependent on a cascade of protein phosphorilations (Collins et al., 1997Collins K. Jacks T. PavIetich N. The cell cycle and cancer.Proc Natl Acad Sci USA. 1997; 94: 2776-2778Crossref PubMed Scopus (283) Google Scholar). Induction of mitosis by IR radiation could thus be due to phosphorilation of cyclins controlling the cell division cycle. The enhanced survival of fibroblasts to UV, induced by IR, could be explained either by a real cellular protective system or by mitosis stimulation; however, stimulation of cell division alone cannot account for the inhibition of the UV-induced cytotoxicity that we observed. Actually, Ara-C abolished non-induced and IR-induced cell mitosis but did not abolish the IR-induced cell protection (Figure 4). The protection is not dependent on neosynthesis of proteins, as it is not abolished when cycloheximide is added to the cells immediately after irradiations (Figure 5). The immediate appearance of the protection after IR irradiation, reaching a maximum 24–48 h later, together with the results of the cycloheximide experiments, suggest that the protection here described is a constitutive mechanism of cellular protection, mediated by molecules existing in small amount in the cells, with further induction by IR. Indeed, if the protection were completely dependent on the neosynthesis of proteins, the cycloheximide would have prevented it. A dose of 250 kJ UVA per m2 decreased the number of cells by 45% (Figure 1) and induced a great amount of lipid peroxidation (Table 2). Because pre-irradiation with IR abolished the UV cytotoxicity but did not interfere with the production of TBARS, we conclude that the IR-induced protection against UVA cytotoxicity is not due to protection of lipids in cell membranes. Abrupt increase in the temperature of the cellular environment induces HSP that protect cells against subsequent challenges from heat (Johnston and Kucey, 1988Johnston R.N. Kucey B.L. Competitive inhibition of hsp70 gene expression causes thermosensitivity.Science. 1988; 242: 1551-1554Crossref PubMed Scopus (264) Google Scholar;Riabowol et al., 1988Riabowol K. Mizzen L.A. Welch W.J. Heat shock is lethal to fibroblasts microinjected with antibodies against hsp 70.Science. 1988; 242: 433-436Crossref PubMed Scopus (535) Google Scholar) or from different cytotoxic agents (Welch and Suhan, 1986Welch W.J. Suhan J.P. Celular and biochemical events in mamammalian cells during and after recovery from physiological stress.J Cell Biol. 1986; 103: 2035-2052Crossref PubMed Scopus (417) Google Scholar). Under our experimental conditions this induction was unlikely to happen. Indeed, we were not able to detect HSP 72 [a highly stress-induced HSP (Welch and Suhan, 1986Welch W.J. Suhan J.P. Celular and biochemical events in mamammalian cells during and after recovery from physiological stress.J Cell Biol. 1986; 103: 2035-2052Crossref PubMed Scopus (417) Google Scholar;Muramatsu et al., 1992Muramatsu T. Tada H. Kobaiashi N. Yamji M. Shirai T. Ohnishi T. Induction of the 72-kD heat shock protein in organ-cultured normal human skin.J Invest Dermatol. 1992; 98: 786-790Crossref PubMed Scopus (58) Google Scholar)] in IR-irradiated cells when we used antibodies specific for this protein, even if increasing temperature was able to induce HSP 72 in these cells (data not shown). Thus, the protection here described is not due to HSP induction; however, we cannot discard the possibility of a mechanism dependent on the localized rise of temperature in the photoreceptor molecules, as suggested (Karu et al., 1991Karu T.I. Tiphlova O.A. Matveyets Y.A. Yartsev A.P. Comparison of the effects of visible femtosecond laser pulses and continuous wave laser radiation of low average intensity on the clonogenicity of Escherichia coli..J Photochem Photobiol B Biol. 1991; 10: 339-344Crossref Scopus (29) Google Scholar,Karu et al., 1994Karu T. Tiphlova O. Esenaliev R. Letokhov V. Two different mechanisms of low intensity laser photobiological effects.J Photochem Photobiol B Biol. 1994; 24: 155-161Crossref Scopus (65) Google Scholar). We do not know the nature of the chromophore mediating the protection here described, but it is worth pointing out that molecules such as cytochrome aa3 (cytochrome oxidase), involved in important cellular biochemical pathways (ATP synthesis, for example), absorb photons in the band between 700 and 900 nm (Rosen, 1978Rosen S. Purification of beef-heart cytochrome C oxidase on octyl-sepharose CL-4B.Biochim Biophys Acta. 1978; 523: 314-320Crossref PubMed Scopus (25) Google Scholar). This kind of absorption could produce vibrationally "hot" molecules (Friedman et al., 1991Friedman H. Lubart R. Laulicht I. A possible explanation of laser-induced stimulation and damage of cell cultures.J Photochem Photobiol B Biol. 1991; 11: 87-95Crossref PubMed Scopus (162) Google Scholar) leading to photochemical reactions or conformational changes. Whatever the molecular mechanisms involved in the IR-induced protection and despite some quantitative variability in the cellular responses, due chiefly to intrinsic differences of cells from different donors, the protection induced by IR radiation described here seems to be a very effective mechanism of cellular defense against the cytotoxicity of solar UV, selected by and conserved through evolution, and thus of major importance for life protection. This mechanism needs to be taken into account, because anthropogenic activities are changing the environment drastically, unbalancing the equilibrium between damage and repair. For example, destruction of the ozone layer leads to an increase of the UVB fluence and, most importantly, to a shift of the UV spectrum reaching the Earth's surface towards shorter wavelengths. Actually, every day, in the natural environment, the cells are first irradiated by solar visible light-IR wavelengths, due to the combined effects of the solar zenith angle and the absorbance properties of atmospheric components. This previous IR irradiation prepares the cells to deal with the following UV radiation. Studies to determine whether IR also antagonizes the mutagenic and carcinogenic effects of UVA and UVB are in progress. The authors thank Mrs. Esther d'Utra e Silva for help in preparing the manuscript. This work was supported by grants from INSERM, European Community (BIOTECH CT 960036) and CERIES (Epidermal and Sensory Research and Investigation Center, Paris).

Referência(s)