Conclusion Like UVB radiation, albeit differently and less efficiently, U V A radiation is clearly mutagenic and carcinogenic. In contrast to UVB radiation, the fundamental photochemistry and premutagenic D N A lesions are still not well determined for U V A radiation. Ten to 20% of sunburn and carcinogenic UV doses from sunlight stems from the U V A b a n d y U V A radiation is therefore an important toxic factor in our natural environment. Summary Ultraviolet B and A radiations (respective wavelength ranges 280-315 and 315-400 nm) are present in sunlight at ground level. The ultraviolet radiation does not penetrate any deeper than the skin and has been associated with various types of human skin cancers. The carcinogenicity of UVB radiation is well established experimentally and, to a large extent, understood as a process of direct photochemical damage to DNA from which gene mutations arise. Although U V A is generally far less carcinogenic than UVB radiation, it is present more abundantly in sunlight than UVB radiation (>20 times radiant energy) and can, therefore, contribute appreciably to the carcinogenicity of sunlight. In contrast to UVB, U V A radiation is hardly absorbed by DNA. Hence, the absorption by other molecules (endogenous photosensitizers) becomes more important, thus radicals and, more specifically, reactive oxygen species can be generated that can damage DNA, membranes, and other cellular constituents. These photochemical differences between U V A and UVB radiations are reflected in differences in cellular responses and carcinogenesis. 35 G. K e l f k e n s , F. R. d e Gruijl, a n d J. C. v a n d e r L e u n , Photochem. Photobiol. 52, 819 (1990).
 P h o t o a g i n g - A s s o c i a t e d L a r g e - S c a l e D e l e t i o n s o f Mitochondrial DNA
By M A R K
Introduction One of the main functions of mitochondria is to supply the cell with energy via oxidative phosphorylation, which is carried out by five protein complexes located in the inner mitochondrial membrane. A proton gradient
METHODS IN ENZYMOLOGY, VOL. 319
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved. 0076-6879/00 $30.00
is generated in a multistep process, involving several redox reactions, resuiting in the generation of ATP from ADP and organophosphate. Erroneous oxidative phosphorylation can lead to the generation of reactive oxygen species (ROS), which is why mitochondria are the main site of ROS turnover in the cell. Oxidative phosphorylation is encoded by both nuclear and mitochondrial (mt) DNA. The human mitochondrial DNA is a 16569-bp, circular double-stranded molecule of known sequence. 1 Mutations in mitochondrial DNA have been found in degenerative diseases and are thought to play a key role in the normal aging process 2 as well as photoaging of the skin. 3-5 The mutation frequency of mitochondrial DNA is much higher than that seen in the nuclear genome, partly due to the lack of protective histones and the absence of nucleotide excision repair, 6 but also due to the deleterious effects of ROS. 7 Reactive oxygen species can generate 8-hydroxy(deoxy)guanosine (8-OH-Guo) in mtDNA directly,8 which can lead to strand breaks. 9 However, they can also lead to large-scale deletions of the mtDNA via an indirect mechanism involving strand breaks during replication of the mtDNA, l°'n The analysis of large-scale deletions via singlet oxygen and their detection in vivo and in vitro will be discussed in this article.
Detection of Large-Scale Mitochondrial Deletions by Polymerase Chain Reaction (PCR) Polymerase chain reaction is a quick and highly sensitive method for selective amplification of DNA. Because mtDNA is of low abundance in the cell, the application of PCR facilitated the investigation of mtDNA mutations greatly.
1 S. Anderson, Nature 290, 457 (1981). 2 D. C. Wallace, Science 256, 628 (1992). 3 J.-H. Yang, H.-C. Lee, and Y.-H. Wei, Arch. Dermatol. Res. 287, 641 (1995). 4 M. Berneburg et aL, Photochem. Photobiol. 66, 271 (1997). 5 M. A. Birch-Machin, M. Tindall, R. Turner, F. Haldane, and J. L. Rees, J. Invest. Dermatol. 110, 149 (1998). 6 D. A. Clayton, J. N. Doda, and E. C. Friedberg, Proc. Natl. Acad. Sci. U.S.A. 71, 2777 (1974). 7 F. M. Yakes and B. Van Houten, Proc. Natl. Acad. Sci. U.S.A. 94, 514 (1997). s T. P. Devasagayam, S. Steenken, M. S. Obendorf, W. A. Schulz, and H. Sies, Biochemistry 30, 6283 (1991). 9 H. Sies, W. A. Schulz, and S. Steenken, J. Photochem. Photobiol. B 32, 97 (1996). 10j. M. Shoffner et al., Proc. Natl. Acad. Sci. U.S.A. 86, 7952 (1986) u M. Berneburg, S. Grether-Beck, V. Kiirten, T. Ruzicka, K. Briviba, H. Sies, and J. Krutmann, J. BioL Chem. 274, 15345 (1999).
Extraction of Total Cellular D N A
Conventional DNA extraction methods using phenol/chloroform, ethanol, and cesium chloride gradients may exclude small cytoplasmic circular DNA, resulting in a low yield of mtDNA. Extraction kits designed to yield DNA of around 5-50 kbp, which avoid phenol/chloroform and cesium chloride, give better results and have the added advantage of removing inhibitory substances such as hemoglobin. The QIA Amp Tissue kit (Qiagen, Hilden Germany) gives good yields for cell cultures (fibroblasts and keratinocytes) and tissues (skin, blood). Extractions of total cellular DNA should yield above 0.1/xg, as lower concentrations can lead to false-negative results. Strategy to Detect Large mtDNA Deletions
For detection of large mtDNA deletions, oligonucleotide primers are designed to anneal outside the deletion (Fig. 1). Thus, amplification of a large PCR fragment can be expected for undeleted mtDNA, whereas the existence of a deleted mtDNA molecule reduces the distance between primers, leading to the amplification of a correspondingly shorter PCR product. Because mitochondrial DNA molecules shortened by deletions are amplified more efficiently than undeleted mtDNA, this strategy results in selective amplification of deleted mtDNA. Short-Cycle PCR
To generate large PCR products, the amplification time during PCR cycles needs to be sufficiently long. By keeping the extension time of the PCR short it is possible to manipulate the selectivity of the reaction to OH
arge-scale deletion A1 B1
FIG. 1. Primer positioning for detection of large-scale mtDNA deletions. Primers A and B anneal outside the deletion and primers C in a nonmutated area. Gray line denotes the
deletionandboxesdenotedirectrepeatssurroundingit. OH,originof heavy-strandreplication; OL, originof light-strandreplication.
LARGE-SCALEOELET1ONSOr mtDNA 1st PCR
2 nd P C R (primershift)
FIG. 2. Experimental design for nested PCR.
preferentially reveal mtDNA deletions. Most mtDNA deletions are between 3 and 10 kbp long. If the amplification time is designed too short for the amplification of the wild-type mtDNA, the deleted molecule is amplified selectively. 12 Nested PCR
As discussed later in this section, different tissues require different detection methods. For a specimen with high abundance of mutated mtDNA, primary PCR is often sufficient. However, tissues with a low content of the deletion of interest may require a more sensitive PCR method, which uses nested primers. 13In this method, a second PCR reaction is carried out, using the initial amplification product as a template. A second set of primers is designed to anneal inside the initial PCR fragment (Fig. 2). Nested PCR can be carried out in several ways. 1. A small volume of the primary PCR can be used directly in the secondary PCR (i.e., 2/zl from a 100-tzl PCR assay). This is the simplest way of carrying out nested PCR. Because nucleotides in PCR are not limiting the reaction and because overabundance of nucleotides may inhibit PCR efficiency, for nested PCR, it is recommended that the nucleotide concentration be one-tenth of the first PCR (i.e., 40/xM instead of 400 tzM in primary PCR). 2. Before transferring a primary PCR product, it can be purified by column purification (Geneclean or Qiagen columns). This ensures removal of inhibitory substances. 12 G. A. Cortopassi, D. Shibata, N. W. Soong, and N. Arnheim, Proc. Natl. Acad. Sci. U.S.A. 89, 7370 (1992). 13S. Ikebe et aL, Biochem. Biophys. Res. Commun. 170, 1044 (1990).
TABLE I OLIGONUCLEOTIDE PRIMERS TO DETECT THE MOST PREVALENT LARGE-SCALE m t D N A DELETIONSa
Ll16 L173 H594 H617 L729 L790 H1390 H1363 L825 L834 H1620 H1641 L219 H447
AACTCAAAGGACCTGGCGGT TFCATGATIq'GAGAAGCCTT AACCCCCTGAAGCTTCACCG CGTGAAATCAATATCCCGCA GCAGTAATATTAATAATTT~CATG TGAACCTACGAGTACACCGA CTAGGGTAGAATCCGAGTATGTI'G GGGGAAGCGAGGTrGACCTG GCCCGTATITACCCTATAGC ACCAACACCTCTITACAGTG GTTGAGGGTFGATFGCTGTAC TGCGGGATA'ITGATI'TCACG ATGCTTGTAGGACATAATAA AGTGGGAGGGGAAAATAATA
1161-1180 1731-1750 5941-5960 6171-6190 7293-7316 7901-7920 13905-13928 13631-13650 8251-8270 8344-8363 16208-16228 16411-16430 219-238 447-466
Annealing temperature (°C)
Site of ±
Size of A
50 55 55 50 58 68 58 68 55 56 56 55 55 55
1836-5447 1836-5447 1836-5447 1836-5447 8469-13447 8469-13447 8469-13447 8469-13447 8648-16085 8648-16085 8648-16085 8648-16085 N/A N/A
3610 bp 3610 bp 3610 bp 3610 bp 4977 bp 4977 bp 4977 bp 4977 gp 7436 bp 7436 bp 7436 bp 7436 bp N/A N/A
aAnnealing temperatures are values, which gave the best results in our PCR assays. Amplifiable deletions are as cited in Kogelnik et al. 14 Abbreviations: A, deletion; L, light strand; H, heavy strand.
Before transfer, the primary PCR is separated electrophretically, and the area of the gel corresponding to the PCR product is excised and processed by a purification technique such as Geneclean before the subsequent nested PCR. This procedure is preferable because it both eliminates any inhibitory substances and results in the transfer of less nonspecific products to the secondary PCR reaction.
Oligonucleotide Primers The human mitochondrial genome is sequenced completely 1 and oligonucleotide primers can be chosen to fit experimental designs. Table I shows primers that amplify the most frequent large-scale deletions, as well as a control region of the mtDNA in which no large-scale deletions have thus far been reported. 14 The primer sequence is given in a 5' to 3' direction together with site, length, and annealing temperature of primer. Primer concentrations around 1/xM produce good results.
14 A. M. Kogelnik, M. T. Lott, M. D. Brown, S. B. Navathe, and D. C. Wallace, N u c l e i c Acids Res. 26, 112 (1998).
Quantification of mtDNA Deletions A PCR fragment from a nonmutated region of the mtDNA can be used as an internal standard for quantification. This fragment should be small so that it will amplify from both normal and mutated mtDNA molecules in order to represent the entire population of mtDNA molecules. In this way, PCR products representing deleted mtDNA molecules can be normalized to the total amount of mitochondria simply by comparing the quantity of deleted mtDNA with that from the internal standard. The number of DNA copies per PCR cycle reaches a plateau toward the end of the reaction. Thus, for quantification it is important that products are produced only during the linear phase of the PCR. This needs to be determined empirically for each primer pair. For this, different concentrations of the template DNA need to be amplified with increasing cycle numbers. The obtained PCR products can be quantitated semiquantitatively by optically comparing dilutions of the PCR with known standards. 3 Furthermore, electrophoretically separated products can be visualized and quantified with a phosphorimager. 5The most sensitive method is the quantification of PCR products via chromatographic separation and subsequent detection with an on-line ultraviolet spectrophotometer (Gynkotek, Germering, Germany). This is especially useful for results where signal intensity is low. Quantification o f P C R Products by Ion-Exchange Chromatographf 5
PCR products for normal and mutated mtDNA are taken up by a sampling robot (20/zl per sample). The negatively charged DNA fragments are then bound to an ion-exchange column and subsequently exposed to an increasing NaCI concentration. This leads to elution according to the total amount of negative charge associated with the DNA fragments (a measure of length). These fractions are then evaluated by an on-line ultraviolet spectrophotometer at a wavelength of 260 nm. The results can be expressed graphically by plotting ultraviolet (UV) absorption (DNA concentration) on the y axis and elution time (DNA amount) on the x axis for each PCR product. Values for the area under the curve of each PCR product can then be used for quantitative comparison. Material Each eukaryotic cell contains several hundred mtDNA molecules with normal and mutated molecules coexisting in varying ratios. 16This so-called 15M. Grewe, K. Gyufko,E. SchOpf,and J. Krutmann,Lancet 343, 25 (1994). 16C. Richter,Int. J. Biochem. Cell Biol. 27, 647 (1995).
TABLE II LEVEL OF LARGE-SCALE DELETIONS IN DIFFERENT HUMAN TISSUES
Level of deletion
Liver Kidney Spleen Prostate Fatty (adipose) tissue
Skin Bone marrow blood
heteroplasmy influences the phenotype of the cell and must be taken into consideration during experimental design. Furthermore, the majority of D N A in a cell is nuclear and m t D N A is only a small and varying fraction of this. Additionally, selection mechanisms have been described that influence the ratio of mutated and wild-type m t D N A molecules on the molecular 2 and cellular level. 17
Large-scale deletions of m t D N A have been detected in many human tissues. Postreplicative tissues, such as muscle and nerve, contain the highest levels of large-scale deletions of the m t D N A (Table II). A second group of tissues, such as liver, kidney, and spleen, exhibit an intermediate level of deletions. Replicative tissues, such as blood and skin, contain the lowest levels of deletions. Highly sensitive methods need to be employed to detect deletions of m t D N A in such tissues. Normal human skin is among tissues with a low content of large-scale m t D N A deletions. It has been shown, however, that sun-exposed skin has a higher content of deletions than sun-protected skin, 3-5 indicating that mutations of m t D N A may play a role in the process of photoaging. Furthermore, it has been shown that singlet oxygen can induce the most prevalent deletion of the mtDNA, u To investigate m t D N A mutations in human skin, punch biopsies are taken from the gluteal region (sun protected) as well 17T. Bourgeron, D. Chretien, A. Rotig, A. Munnich, and P. Rustin, Z Biol. Chem. 2d18, 19369 (1993).
as from the neck and forearm (sun exposed). Specimens (4-6 mm diameter) are subjected directly to extraction of total cellular D N A or can be first snap frozen in liqid nitrogen and stored at - 8 0 ° until use. For extraction of D N A from skin, incubation times with proteinase K may have to be increased from the normal 1-2 hr up to 6 hr to break down architectural structures of the skin. Blood samples can also be employed for the detection of large mitochondrial deletions. However, anticoagulants should not contain heparin as this may inhibit subsequent PCR reactions. In Vitro
Deletions of the m t D N A can be detected in cell culture. The conditions of culture, treatment, and assay of choice for the detection of mutations vary from cell to cell. Normal H u m a n Fibroblasts Fibroblasts are involved in photoaging--changes of the skin after chronic exposure to ultraviolet light--in which reactive oxygen species play a critical roleJ s It has been shown that the most prevalent large-scale m t D N A deletion, also called common deletion, can be detected in normal human fibroblasts and that the abundance of this deletion increases after repetitive irradiation with U V A light. Evidence suggests that this increase is due to the effects of singlet oxygen, u There is no nucleotide excision repair in mitochondria. 6 However, m t D N A is repaired readily after ROSinduced damage after 4 hrJ 9 In order to shift the steady state in cells between ROS-induced damage and ongoing repair toward the damage side, they either have to be exposed to ROS-generating systems continuously or at least in shorter intervals than the time necessary to remove generated damage. Furthermore, when irradiating cells repetitively, the lethal effects of U V light have to be taken into account. Figure 3 shows the effect on cell viability of different U V A doses. In this experiment, cells were exposed to doses of 0, 4, 8, and 16 J/cm 2 of U V A three times daily. Irradiation with 16 J/cm 2 three times daily induced significant cell death, whereas 8 and 4 J/cm 2 three times daily had little effect. To induce the common deletion, normal human fibroblasts can be irradiated with 8 J/cm / at 4-hr intervals. Irradiation with 4 J/cm 2 also induced the common deletion, but the total time needed for induction is 18K. Scharffetter-Kochanek,Adv. PharmacoL 38, 639 (1997). 19W. J. Driggers, S. P. LeDoux, and G. L. Wilson, J. Biol. Chem. 268, 22042 (1993).
Number of Irradiations
FIG. 3. Viability of normal human fibroblasts after repetitive UVA irradiation. Normal human fibroblasts were irradiated for 12, 24, and 36 times at 0 (,), 4 (B), 8 (A), and 16 (e) J/cm2. After each interval, cell viability was assessed by trypan blue exclusion before and after irradiation and compared to sham-irradiated cells.
longer. These doses are of physiological relevance as they are similar to those administered to h u m a n skin during the course of a 15- to 30-min exposure on a s u m m e r day at noon with a northern latitude of 30-350. 20 Furthermore, repetitive irradiation of cells is comparable to the type of sun exposure received by tourists on s u m m e r holidays.
Culture and Medium N o r m a l h u m a n fibroblasts are grown in 10-cm petri dishes in Eagle's minimum essential m e d i u m (Life Technologies G m b H , Eggenstein, Germany) containing 15% fetal calf serum (Greiner, Frickenhausen, Germany), 0.1% L-glutamine, 2.5% N a H C O 3 , and 1% streptomycin/amphotericin B in a humidified atmosphere containing 5% COe. For irradiation, the m e d i u m is r e m o v e d and replaced with phosphate-buffered saline (PBS).
UVA Irradiation Lids of petri dishes are r e m o v e d and cells are irradiated with a U V A SUN 5000 B i o m e d irradiation device (Mutzhas, Munich, G e r m a n y ) whose emission is filtered with U V A C R Y L (Mutzhas) and U G 1 (Schott Glaswerke, Munich) so that it consists of wavelengths greater than 340 nm. Irradiation is carried out three times daily for four consecutive days and cells are checked for viability by trypan blue exclusion. To investigate the 20j. E. Frederick and A. D. Alberts, "Biological Responses to Ultraviolet A Radiation" (F. Urbach, ed.). Valdenmar, Overland Park, KS, 1992.
effects of repetitive irradiation longer than 4 days, cells are subsequently aliquoted 1 : 1 with one aliquot stored at -80 ° until extraction of DNA and the other aliquot is plated to a new 10-cm culture dish for ongoing culture and irradiation. In vitro irradiation of normal human fibroblasts with this regimen induces the common deletion to detectable levels after 24 irradiations with a maximal induction after 36 exposures.
Coincubation with Deuterium Oxide (DeO) Deuterium oxide enhances the half-life of singlet oxygen. Coincubation of fibroblasts with D20 during irradiation leads to a small but consistent increase of the common deletion when compared to UV irradiation alone. The high toxicity of D20, which is observed when cells are exposed repeatedly, requires that cells are only exposed to D20 during irradiation with D20 (99.9 atom % 2H) at a final concentration of 95%.
Incubation with NDPOe Incubation of cells with a singlet oxygen-generating system can also induce large-scale deletions of the mtDNA. The endoperoxide of the disodium salt of 3,3'-(1,4-naphthylidene)dipropionate (NDPO2) generates singlet oxygen through thermal decomposition. Its generation, biochemical characteristics, and applications are discussed in more detail in a different section of this volume. This singlet oxygen-generating system is well suited for cell culture applications due to its good water solubility and absence of toxic effects at concentrations up to 40 mM for a 1-hr incubation period. Incubation of fibroblasts with 0.3 mM NDPO2 in PBS for 1 hr in the dark at 37°, instead of UV irradiation, leads to the induction of the common deletion with kinetics comparable to UV irradiation.
Coincubation with Quenchers of Singlet Oxygen The singlet oxygen-mediated induction of the common deletion can be inhibited by coincubation of fibroblasts with substances that quench the effect of singlet oxygen such as sodium azide (NAN3) and vitamin E (c~tocopherol succinate). Sodium azide should be applied in concentrations of 20 to 50 txM. The presence of NaN3 leads to a dose-dependent reduction of the irradiation-induced generation of the common deletion. To ensure uptake of vitamin E into cells it has to be in culture 24 hr prior to the first irradiation. Concentrations of 25/xM vitamin E show good results in our system.
Keratinocytes It has been shown in vivo that the epidermis has a lower content of the common deletion than dermis. 5 This may be due to the fact that keratinocytes, the predominant cell type in the epidermis, have a higher efficiency in quenching ROS when compared with fibroblasts, which prevail in the dermis. Nevertheless, the repetitive irradiation of keratinocytes does induce the common deletion, but it does require higher doses or longer irradiation periods.
Summary Heteroplasmy, replicative segregation, low copy numbers of mtDNA, and selection mechanisms at the molecular and cellular level are all factors that determine requirements toward the experimental design for the detection and the quantification of mtDNA mutations. The short half-life and low stability of ROS further increase the technical demands. However, the continuous improvement of techniques has given us more insight into the interactions between ROS and mtDNA, both at the level of endogenous ROS produced by the normal mitochondrial metabolism and exogenous sources of ROS, such as singlet oxygen, which can result from treatments such as U V A exposure.
Role of Activated Oxygen Species in Photodynamic Therapy SHARMAN, CYNTHIA
and JOHAN E.
Introduction Photodynamic therapy (PDT) finds its roots at the turn of the century when a young medical student found that acridine orange killed paramecia on exposure to sunlight. 1 Subsequently, a plethora of information concerning the lethal effects of the combination of photosensitizers and light, both in vitro and in vivo, has been documented. Several reviews have been published outlining the clinical aspects of PDT and the preclinical and clinical studies predominantly accomplished within the past 25-year pe-
1 0 . Raab, Z. Biol. 34, 524 (1900).
METHODS IN ENZYMOLOGY, VOL. 319
Copyright © 2000 by AcademicPress All rightsof reproductionin any form reserved. 0076-6879/00 $30.00