Photostability of DNA

Abstract

Ultra-Violet (UV) radiation may result in skin cancer due to the hypothesis that DNA photostability provides a small allowance for photoreactions. DNA has been protected by the Ozone layer from the highly genotoxic UVC wavelengths but allows UVA and UVB wavelengths to irradiate to the earth’s surface. At first, these two wavelength categories may seem harmless as there is no immediate harm initially, but after prolonged exposure, due to sunlight and other factors, there has been correlation between UVB and UVA with cancer. AIMS After irradiation, DNA forms excited states of which can be analysed and their decay channels determined using a variety of experimental methods such as ‘time-resolved spectroscopy’ and femtosecond fluorescent pump-probes, to name a few. There are a variety of photo-products that can arise such as REACTIONS Cyclobutane Pyrimidine Dimers, Pyrimidine(6-4)Pyrimidone Dimers and oxidatively generated photoproducts, which are derived after the formation of Reactive Oxygen Species (ROS). For a small portion of the products that form the body already has mechanisms in place, by using enzymes, to reverse the lesions that cause mutations, but this is not widespread for all photoproducts. After compiling information from experimental resources and reviews and can be concluded that DNA is already photostable as there are exemplary mechanisms already in place to rectify the chance of dangerous lesions. In light of this, research should focus on organisms with more advanced photo-protection mechanisms to innovate treatment for the rare occasion in which a photoinduced mutation is not rectified.

Glossary and abbreviations

• 6-4PPs : pyrimidine (6-4) pyrimidone
• c,s Thy<>Thy : cis-syn isomer of cyclobutane
• CPD: Cyclobutane Pyrimidine Dimer
• DNA : Deoxyribonucleic Acid
• EMS : Electromagnetic Spectrum
• FADH^– : reduced Flavin adenine dinucleotide
• fs : Femtoseconds
• HPLC-MS/MS : High Performance Liquid Company tandem Mass Spectrometry
• M^-1s^-1 : per Molar concentration per second
• NADP: Nicotinamide Adenine Dinucleotide Phosphate
• Nm = nanometre
• ¹O₂ = Molecular Oxygen in its Singlet excited state
• ^·OH : Hydroxide radical
• Pyr<>Pyr : Cyclobutadiene pyrimidine dimers
• ps : Picosecond
• ROS : Reactive Oxygen Species
• T<>T: Thymine cyclobutadiene pyrimidine dimers
• UV : Ultra Violet
• WC : Watson and Crick theory (of DNA)

1.    Introduction

Long before the Watson-Crick model for DNA hypothesis was accepted, PG Unna proposed that Ultraviolet radiation (UV) cause skin cancer in his article ‘The Histopathology of the Skin’¹. It is only suitable that there is a large interest in the topic of UV damage and a large array of studies to prove that there is a correlation between cellular sun exposure and skin cancer². Solar radiation from the sun is the first provider of UV exposure, and its components that reach the earth’s surface have been determined as class I carcinogens³. The reason for this is because they bring about DNA damage by inducing mutations on the p53 tumour suppressor gene as well as generating oncogenes⁴. It should be remembered that UV does have benefits to human health. An example of this being that it is the primary contributor to Vitamin D production, which can be used to protect from various diseases such as Multiple Sclerosis and Heart Disease⁵.

Mutations frequently occur as a phenomenon that can be reversed by repair mechanisms specific to DNA to conserve genetic integrity⁴. Damage and alterations to the genetic code are eradicated and replaced to correct the sequence back to its original form using nucleotide and base excision repair pathways⁶. On occasion, these corrective mechanisms may not return the sequence entirely back to its original code as the bases may have been misread, meaning that the mutation is not corrected⁴. Because of this, vital proteins and messages are not correctly translated and, in the case of cancer when the mutation is on a tumour suppressor gene, a growth is allowed to form creating fatal consequences, which are hard to reverse at times.

Photons in light can provide enough energy to break bonds in organic molecules, and this applies to UV. This is of relevance as the structure of DNA, bases and sugar backbone moieties alike can be under threat. Hence, the need to understand natural defence mechanisms help organisms deal with the threat is necessary. By following the mechanisms of photoreactions and understanding DNA excitation, it should be possible to realise to what extent DNA is a target of UV and to investigate how the products, as lesions in DNA, can cause cancer in skin cells. Damage to DNA is brought about by either direct absorption of UV photons or photosensitization mechanisms.  Although there is a broad spectrum of reactions which can occur with a vast array of photoproducts, this project only plans to detail the main mechanisms and products which bring about damage.

UV is a spectrum of wavelengths, ranging from 200 to 400nm in the electromagnetic spectrum of wavelengths, which can be conveniently broken up into three main categories; UVA, UVB and UVC. UVC is the shortest wavelength, occurs in the wavelengths <280nm and it is the most cytotoxic due to its higher energy in comparison to the other two⁸. Solar radiation consists of all three categories, but we do not witness the effect of UVC on the earth’s surface due to the presence of Ozone and Oxygen in the atmosphere, which acts as a protective barrier. Therefore it is not relevant to consider its effects on DNA for this project as a threat is not imminent. As evident in Figure 1, Ozone absorbs all UVC wavelengths, as does DNA. This provides support for the idea that ozone protects us from the immediate threats of UV, as without it, DNA would be a providing a tremendous threat to living organisms. It is for this reason that scientific focus has centred around UVB, and of recent, UVA as they are the most relevant.

UVB occurs between wavelengths 280-320nm it is the main contributor to Vitamin D production, although its incidence in the atmosphere is only 10% percent as it is also efficiently absorbed by Ozone and Oxygen^9,10. It is the most Carcinogenic of the wavelengths in the spectrum that reaches the earth’s surface and for many years has been thought the main source of Photolesions in DNA¹⁰. UVA occurs at 320-390nm is supposedly less harmful than UVB. Although it may not be as carcinogenic, is not entirely harmless as there is growing evidence for its involvement in both direct absorption and photosensitization mechanisms of DNA⁹. Although the scientific community has accepted its involvement in participating in the formation of skin cancer, the exact mechanisms and pathways it provides are still unclear. But it has been agreed that the main damage that occurs via UVA is by photosensitisation mechanisms².By studying UV’s consequences on cellular DNA and single bases, it can be evaluated that there are two main reactions that occur¹¹. The first is the direct absorption of UV photons by DNA which results in the formation of photoproducts by photoinduced reactions¹². The second is the absorption of UV photons by endogenous species leading to the generation of electronically excited sensitizers, which can result in the occurrence of photosensitization mechanisms.

DNA molecules are suitable chromophores of most UV wavelengths¹³ (Figure 1), and over the last ten years, there has been a rise in the topic around the excited electronic states of DNA, by which these reactions can occur. Many studies were done on a wide range of cultured cells using a variety of methods, such as Time Resolved Spectroscopy¹⁴, to understand how DNA Photostability can arise while considering the contribution of DNA structure. It is important to note that it is not the photostability of DNA that is in question as if it were not photostable, most organisms would not be alive. What is in question is the precise mechanisms involved in ensuring the photostability of our DNA, the repair mechanisms and the control of mutation formation.

2.    Discussion

2.1.           The Structure of DNA

We already know that DNA is one of the most fundamental building blocks of most life forms. Even the slightest fault in it, if not rectified, can be fatal. The main structure consists of 4 nitrogenous bases; Adenine, Cytosine, Thymine and Guanine. In 1953, Watson and Crick presented a structure to their peers, and it is this model that we still use to this day. They proposed that DNA consists of a double helical structure¹⁵, held together by hydrogen bonds¹⁶, paired up according to specific base pairing rules; A-T, C-G (Figure 2).  The hydrogen bonds arise from dipole-dipole interactions in hydrogen atoms, which acts as a bridge between the electronegative Nitrogen and Oxygen atoms (Figure 2).

A phosphate group joins nucleotides together by phosphodiester bonds and is attached to a deoxyribose sugar, which is then connected to the nucleotide by a ß-N-glycosidic linkage (Figure 2). Together, these make up the phosphate backbone that holds the strands in place. The molecular geometry of all the nucleotides is the same, which allows for symmetry and base stacking. Base stacking provides stability, along with base pairing, of the DNA molecule and consists of noncovalent interactions between aromatic rings in the DNA molecule that is known as p-p stacking.

2.1.1.      π-π Stacking

In a double helix, the intrastrand distance is around 3.9Å which is very close to the range in which Van der Waals Interactions occur¹⁷. This means that π-π interactions can occur to organise and stabilise the structure of DNA. In DNA,  π-π interactions are parallel as this orientation increases the amount of H-Bonds, which research suggests is made even greater when electron donating substituents are bonded to the aromatic ring¹⁸. Electron withdrawing substituents increase the negative charge available to the ring which creates a repulsive force between the stacked bases. In DNA molecules, the concepts of p-p stacking and Hydrogen bonding run in coordination, but there is not enough information on how they interconnect, which is also reflected in attempts to try and study their influence on excited state dynamics. A well-established hypothesis states that DNA lesions are formed via the singlet excited states, and it is hard to pinpoint which of the two interactions contribute the most to the formation and dynamics of these excited states¹⁹. This is because hydrogen bonds allow for the transfer of protons and possibly protons between base pairs whereas p-p stacking in the planar aromatic bases helps to form excited state dimers due to two bases in close proximity ‘sharing the excition’¹⁹. This inconclusive explanation is not sufficient enough to apply to the formation and nature of excited states; therefore a more in-depth look which looks at excitation mechanisms and different methods of decay is required.

2.2.        DNA Excited States

As highlighted in the introduction, the importance of studying the excited states that arise after UV exposure can help to understand DNA Photostability. They provide useful information on the deactivation pathways, which may occur as well as the states in which there is a window of opportunity for the occurrence of photoreactions. All dynamics induced by UV absorption in monomeric DNA bases have been widely reviewed and most reviewers agree on the fact that these dynamics are brought about by photons in the ¹pp^* excited state^20–22. Previously, fluorescent spectra²³ and Femtosecond UV/Visible light probes²⁴ were used on single bases to show that there are low quantum yields during fluorescence experiments. A quantum yield is the probability of an excited state being made inactive by fluorescence, as opposed to non-radiative mechanisms²⁵.  By correlating the information in reviews to studies that analyse absorption spectra of nucleotides it can be implied that these low yields can be derived from non-radiative decay process. These processes have also experimentally proven to absorb intense UV at 260nm  which exhausts the ¹pp^* state transition²⁶. In single bases it has become important that once this excited state is reached, internal conversion occurs to return the base back to its ground state. In experiments, this is proven by a far larger signal magnitude than that given by oglionucleotides¹⁹.

Internal Conversion is a radioactive decay procedure where the excited nucleus of atoms interact electromagnetically with one of the orbital electrons of the same atom, which allows for the emission of a photon. As internal conversion occurs in single bases, vibrationally warm molecules are created and the system returns to its electronic ground state, as demonstrated in Fig.3. In the electronic ground state, vibrational cooling occurs at a time frame of 1-10ps and heat is transferred to the surroundings¹⁷. This means that any excited vibrations decrease in energy. The lifetimes of the bases in the ¹pp^* state is a sharp contrast to other heterocyclic compounds they are related to, allowing evolutionary visionaries to believe that this property is interconnected to the selection pressures that helped this characteristic to evolve²⁷.

The first accurate measurement of the ¹pp^* was achieved by Pecourt et al, using femtosecond transient absorption spectra²⁹. This proved to be a breakthrough as they hypothesised that ultrafast movements occurred between excited electronic states when a wave function comes into contact with a concial intersection. Conical intersections are degenerate points between two potential energy surfaces at which any non-adiabatic couplings are always present.  Theoretical knowledge may be proficient to compare the DNA bases but difficulties occur because of nonadiabatic interactions that occur between excited states.  Because of this, scientists researching base Photostability have turned to other methods, such as time resolved spectra, to supplement their theoretical findings^28,31 as well as using pioneering computational methods³². Computational methods are highly efficient and quick to perform which allowed Ismail et al to confirm that decay pathways between the Franck-Codon regions in the base Cytosine that had no boundaries and achieved the ground state by conical intersections. Conical intersections can now be found for all DNA bases and can provide supplementary information for DNA Photostability (discussion to follow using Figure 4).

Time resolved spectra is an ideal method as it provides much detail and can characterise internal conversion mechanisms. From these experiments, it has been made clear that ring puckering could be the key to help deactivate bases, making them photostable^33,34.  It is not possible to describe photodynamics using just the pathways depicted in Figure 3 therefore; Dynamics Simulations can provide a new perspective and help to paint a more realistic description. Ab initio photodynamical simulations have been performed by researchers in varied conditions for each base^35–37. These simulations aimed was to utilise the nonadiabatic trajectory dynamics simulations on Guanine, Cytosine and Uracil bases to come up with suggestions for excited state dynamics, along with the help of the revalidation on the experiments of Adenine and Thymine^37,38. This can produce a profuse amount of information in which can be compressed into a more general view which also takes into account the dynamic simulations and the principles of the Photostability of bases³⁹. From these experiments, there are several, clear conical intersections which become known.

These conical intersections, along with the help of symbolic notation, can be explained. Figure 4 provides a good representation of the conical intersections that arise, as proposed by Barabtti et al³⁹ that can arise in the Thymine, Adenine and Cytosine bases. A-D represents the typical geometric distortions that occur and E-G provides characteristic examples of these geometries. The thicker lines depict the molecular ring plane and the thinner lines connect the atoms, which have been moved out of the plane, due to distortion. Barbatti et al concluded that the main way conical intersections are reached, between the ground state (S₀) and the first excited state (S₁), is to rotate/twist one double bond by 90°. By doing this, a bi radical state is created⁴⁰. Further to this, by adding one extra degree of freedom. The energy gap between the two is reduced which ‘tunes’ the intersections³⁹. In heterocyclic compounds, such as DNA bases, puckered structures form because of this. Using the example of Adenine (E in Figure 4), the CH₂ group has been moved out of the plane of the heterocycle. This means that it is twisted compared to the first N atom(N₁) and describes the crossing by the pp^* and closed shell state³⁹.

It is safe to assume that as single bases, there is photostability by the decrease in the probability of a photoreaction occurring due to ultrafast, non-reactive deactivation of excited states but in stacked bases it gets more complicated. In stacked DNA, the four bases are held relatively close to one another. The ¹pp^* states are also spatially and energetically in close proximities which delocalizes the excitation over a few bases, or base pairs^20,22 and it is for this reason the photo physics of native DNA becomes complicated to explore. As Figure 3 suggests, the decay routes the excited states may take are very similar to that of single, bases without stacking. However, in stacked bases there are additional decay pathways and reaction routes which become imperative due to interactions that occur between nucleotides.

DNA in its native state is hard to record with spectra, as several issues arise. There is selective excitation for a specific type of base in the DNA sequence and there is an overlap of absorption bands between the four bases. It is for these reasons that electronic states of the bases are used to explain how photoproducts may arise. There are a variety of photoproducts that can form due to UV exposure but only a few can be detailed here. The emergence of time resolved experiments have helped to investigate the kinetics involved in DNA Photochemistry. They have taken the issues presented into account due to the large time range of the experiments, covering a time span in the femtosecond (fs) to millisecond (ms) range. Bucher et al did a time resolved study of double stranded DNA and was able to come up with several conclusions using their transient spectra of three different time scales: 5ps range, 40ps range, 210ps range. The 5ps range showed ‘signatures’ of the vibrational cooling of DNA molecules as well as evidence of the fast decay of the excited states to the ground state⁴¹.

Although promising, the results did show a few differences in the kinetics of which several researchers have proposed models to demonstrate them. The first would be that proposed by Domcke et al^42–44, which used the idea that a process called ‘direct coupling’ excites the single bases which causes an interchain charge transfer, facilitated by base stacking. This results in an ultrafast motion along a proton transfer co-ordinate, and this can explain the ultrafast decay that single bases exhibit when they are excited by UV photons.  Bern Kohler et al^45,46on the other hand, proposed a different model where excitation and charge transfer occurs within a single strand. This would suggest that the ultrafast decay is influenced by the hydrogen bonds.  There is a third possibility that Takeuchi et al⁴⁷ have researched where an exciton induces a double proton transfer between the hydrogen bonds of the paired bases.  16 years ago Kang et al⁴⁸, successfully recorded the excited state lifetimes of Thymine, which returns to the ground state in the time frame of 5-7ps. This fact is the only agreement when considering DNA photophysics, especially thymine which forms most dimeric lesions. As evident above, each model explains the results from their own individual finding, which makes it particularly difficult to try to pinpoint which is the most probable as the methods are yet to be compared. As a result of this, the understanding of the dynamics of excited bases are not clearly understood, and it would be beneficial for more experiments to be performed so that the molecular structure of the different states can be understood and precise decay mechanisms can be proposed. The information obtained would be a significant contribution in understanding the different excited states of DNA in its native form as well as further knowledge of the lesions that can form.

As mentioned earlier, the intrastrand distance between bases is around 3.9Å which indicates that light induced reactions are possible, creating Photolesions from the excited singlet, ¹np^* or triplet states. In some cases, the excited complexes that have a strong charge transfer character may have the possibility of proton transfer occurring across Hydrogen bonds between paired bases. This leads to the occurrence of excited energy levels that can occur⁴². Excited energy levels can occur from these processes, which have approximately 50% occupation¹⁷. Can this perhaps play a part in the formation of lesions and damage repair?

The generation of excited states allow us to monitor decay processes that occur, which provide key facts in understanding DNA Photostability. Due to the ultrafast decay processes that occur, in most UV irradiation occurrences, reactions cannot take place as excited complex return to their ground state. Although this is the case, there are occasions where longer lifetimes may occur and this could be an ‘escape route’ making it easier for photoreactions to occur.

2.3.           Types of Photoproducts

DNA damage, with relation to UV may be brought about by the following:

• Direct Absorption –  UV Photons are absorbed by  the cellular material to their excited states which can result in a photo induced chemical reaction^12,49. A common product has been pyrimidine dimers⁵⁰.
• Photosensitization Mechanisms – UV Light is absorbed by endogenous or exogenous photosensitizers that are excited to their triplet-excited states and can harmlessly decay to their original ground state via intramolecular decay processes. Photosensitization mechanisms can be split into two different types^12,49 which will be later outlined. Products of these two types of mechanisms may react with components of DNA which can lead to lesions within the molecule as they can modify the bases and disrupt or cause intra- and inter-strand crosslinks⁵¹.

Four different dimers that can arise as a result of direct absorption of UV photons: Cyclobutane Pyrimidine Dimer’s (CPDS), (6-4) pyrimidine pyrimidone dimers (6-4) PPs, Dewar-valence isomer dimers and spore dimers. CPD’s are the most common and absorb spectra at shorter wavelengths to DNA bases²⁸.

2.3.1.      Pyrimidine Photoproducts

Pyrimidine bases ( Thymine and Cytosine) are implied to have excited states with ¹np^* character which can have longer lifetimes²².  In these bases, intersystem crossing transitions to the triplet state with ³pp^* character⁵²(Figure 3). Experiments can confirm that UV light of 240-280nm excites Cytosine and Thymine to a higher singlet state (S₁) of which as a lifetime on the picosecond scale before it either decays or transfers to the triplet excited state via intersectional crossing⁷. When exposed to UV radiation of 260nm-300nm, a [2+2] cycloaddition reaction occurs and cyclobutane pyrimidine dimers (CPD’s) form for both bases. Another important pyrimidine photoproduct is pyrimidine(6-4)pyrimidines (6-4PPs), which a [2+2] addition occurs when a pyrimidine base in its singlet state reacts with another pyrimidine base¹².

2.3.1.1.              Cyclobutane Pyrimidine Dimers (CPD’s)

Cyclobutane Pyrimidine Dimers are notorious for causing the most significant damage to both cellular DNA and model systems after exposure to UV light⁵³. They form via a cycloaddition reaction between a C5=C6 double bond between two pyrimidines which results in the formation of a 4-membered ring to link the bases together^54,11(Fig. 4). This process often happens after excitation between two adjacent Pyrimidine bases in the same strand, but it is not uncommon for it to occur on two opposite bases¹². There is wide speculation over the mechanism by which CPD’s form with two clear theories arising. The first is that these photoproducts form from their Singlet excited states, and this is the most consistent theory^55–57. The second is that Triplet excited state may also play a part¹³ but when studies have been done to investigate this, the conclusion rules that the triplet state plays no role, or insufficient evidence does not conclude the matter^56,57.

There are CPDs for all combinations of pyrimidine dimers but Thymine<>Thymine dimers are the most common lesion and this has been confirmed by using HPLC-MS/MS^58,59. This technique utilizes the basic principles of polarity to separate products in a solution, and records their mass to charge ratio using a spectrometer which helps us to identify the compounds⁶⁰.  There are four possible isomers (Figure 5.) and the cis-syn isomer is the major product formed when DNA reacts with UV^61,7.  In native DNA the trans-syn is the least common diastereomer due the steric constraints that arise due to the sugar phosphate backbone⁶².

CPD’s which have Cytosine could possibly undergo a reaction in which the C5-C6 bond can undertake a deamination reaction where the C4 amino group is hydrolysed to give a carbonyl substituent⁶³(Figure 6). Although Uracil Hydrate is more likely to form from a cytosine monomer than a CPD, any product formed will contain Uracil which could be involved in the mutagenicity of the CPD lesion in DNA⁶⁴.

2.3.1.2.              T(6-4)C Dimers

These products are not as common as CPDs but are still an adduct of photoinduced damage to DNA. The dimers form when two bases, in their singlet excited state, are linked by a single bond between two bases, one end of the bond on the C6 position of a heterocycle and the C4 of a carbonyl or imino substituents on Cytosine or Thymine^65,66. The products that result are azetidines and oxetanes which are not stable and quickly reposition to generate (6-4)PPs where carbonyl or imino substituent gets transferred from the 3’end base to the 5’ base onto the C5 position¹².  These lesions have low quantum yields, with the highest yields available for 6-4 dimers with thymine and cytosine⁵⁹. Thymine(6-4)Thymine dimer has a pyrimidine moiety with an absorption band that has a peak at 325nm¹⁴. As Figure 7 shows, the band for Thymine(6-4)Thymine dimer is highly different to the absorption band for calf thymus DNA. This provides a good indication that it is a completely separate product to thymine itself and that a lesion has been formed.

In studies, flash photolysis has been performed on thymine oligonucleotides, (dT)₂₀, when dissolved in a solution of H₂O.The (dT)₂₀ were excited at a wavelengths of 266nm and the band which characterises the pyrimidine moiety rises in a millisecond timeframe. This time is longer than that of decay of the thymine excimer in the triplet state (which occurs within 10ns⁶⁷) or the singlet state ( occurs at approximately 500fs²⁰). This information implies that the (6-4)PP lesion could be formed via the ground state intermediate, which would explain why it is spectroscopically quiet in Visible-Light and near-UV wavelengths.

The 6-4pps are made during a [2+2] cycloaddition between a C5-C6 double bond of a 5’ end base and a C4 carbonyl of a 3’ end base (Figure). This type of reaction between an excited carbonyl and alkene is well known in organic chemistry, and it is known as the Paterno-Buchi Reaction⁶⁸. The products of this reaction are oxetanes or azetidines depending on which base is on the 3’ end. Observations during experiments agree with the suggestions that oxetanes are involved and even imply that it could be that (6-4)PPs are formed via this ground state intermediate ^14,69. Oxetane has a smaller extent of p-conjugation compared to the thymine part of the dimer and when yielded in a reaction, can be isolated due to their long lifetimes. They can undergo a rapid ring opening which may be able to account for the millisecond rise in spectra¹⁴. (6-4)PPs have a different electronic system to its reactants, and it can absorb longer wavelengths, of around 325nm (Figure 7), which happens to be within the UVA region⁷⁰. This implies, as reported by Cadet et al⁶⁵, that sunlight can trigger the rapid conversion of the (6-4)PP photoproduct into its Dewar isomer.

The Dewar isomer formed (shown how in Figure 8) conveys several features which have piqued the interest of researchers. In time-resolved experiments that have been performed excitation has caused peaks to form at approximately 320nm, with the population of the ¹pp^* state⁷⁰. By using Quantum Chemical Modelling, the excited state potential surfaces may be deducted to understand which excited state may lead to the formation of Dewar isomers⁷⁰. It can be deduced that the conical intersections depend heavily on the interactive forces that exist between the heterocycles and to some extent the presence of the phosphate backbone.

 The biological consequence of both (6-4)PP is not yet detailed but some researchers have implied that they are counteracted by photolyases, in a similar manner to CPD’s^71,72. Although unideal when analysing photoproducts, it can convert into its Dewar isomer which may be able to provide valuable insight into how could perform in our genetic fabric if further analysis is carried out on it.

2.3.2.      Rectifying CPD’s

It is a fact that thymine CPD photoproducts ( T<>T) are a common occurrence when DNA is exposed to UVB light⁷³. It is only fitting that the next step into understanding how skin cancer arises from dimers is to know the repair mechanism that has evolved over time by our bodies to rectify the distortion in the DNA structure. Over several studies a group of scientists^74–76, at Ohio State University and the University of North Carolina, have been able to come up with several conclusions concerning the mechanism of repair and the enzyme involved and the reactions that take place.

By consulting papers, such as that by Dr Aziz Sancar^74,76,77, where the function and structure of  DNA Photolyse have been defined. They were able to draw from the idea that the incorporation of a fully reduced Flavin adenine dinucleotide (FADH^–), was so that it could act as an electron donor and catalytic co-factor⁷⁶. They used this to come up with a repair mechanism which has been depicted below in Figure 9.

As Figure 9 shows, FADH^– undergoes excitation to start its repair function. By drawing from previous models of the catalytic reaction A. Sancar et al^75,76, proposed a reaction where the excited FADH^-* transfers an electron to the T<>T which produces a charge separated radical pair (FADH^· + T<>T^·-). The ring of the dimer is now anionic which causes it to split by a [2+2] cycloreversion. The residual electron is returned to the Flavin cofactor radical to restore the negative charge, regenerating FADH^–, which closes the cycle. When the paper was reviewed the mechanism had yet to be proven directly, therefore A.Sancar et al⁷⁷ mapped the repair process by performing femtosecond synchronisation of bacterial DNA Photolyse dynamics. This experiment gave positive results, showing that the photocycle was through a radical mechanism and the process occurred on a subnanosecond timescale⁷⁷. They revisited their concept in a paper six years later where they used the concept that CPD’s can be fully restored by Photolyse in visible blue light⁷⁶ to understand each pathway (indicated K_AB in Figure 9). By using ultrafast UV absorption spectroscopy they were able to come up with a complete cycle of CPD repair and were successful in explaining the molecular mechanism (Figure 10).

Unfortunately, although this may work for T<>T dimers, the repair process may overlook the presence of C<>C dimers, which could be an indication as to what goes wrong and why the cascade of the process that occurs after, can cause skin cancer.

2.3.3.      Where does it go wrong?

A mutation in the p53 gene, which suppresses tumours, is the location of the most genetic defects in Skin Cell Carcinomas (SCC) which is the second most common form of non-melanoma skin cancer and has been shown to be linked to UV after continuous exposure⁷³. The p53 protein is required to stop tumour cells from replicating; therefore if there is a mutation on this gene, it stops the p53 protein from correctly forming its primary, secondary and tertiary structure which in turn hinders its function. When the function of the p53 gene is lost the damaged cells lose the ability to undergo programmed cell death, apoptosis, which allows the cell to replicate causing a tumour to form^78,79. Genes which account for skin carcinogenesis often have a ‘UV signature’ and in p53, the majority are CàT single base transition mutations or to some extent CCàTT tandems which tend to be highly deleterious^80,81.

Jean Cadet and Thierry Douki proposed, and in agreeable experiments, that CPD formation was preferential at locations on the gene where thymine was adjacent to another thymine or a cytosine². They further discovered that although there are CC photoproducts, which are generated in low yields, they display a high mutagenic potential proving its worth as a characteristic signature of UV radiation⁵⁸. Identifying these products in experiments can prove difficult as there is a deamination reaction, which can occur, where cytosine is converted into uracil, (Figure 6).

2.3.4.      Purine Photoproducts

Purine bases have assumed to be a target of UVB radiation but to a much lower extent. Intrinsically speaking, Adenine and Guanine are far more photostable than the pyrimidine bases as they transfer the photochemical excitation energy to neighbouring pyrimidines⁵⁶. In Rahn’s “Search for an adenine photoproduct in DNA”⁵⁶ he concluded after his research that the photoproduct of adenine was inhibited from forming by the fact that it was held to thymine by Hydrogen bonds. He further suggested that the interactions of Purines with other molecules in DNA traps the energy required to form photoproducts, therefore they do not occur⁵⁶. Secondly, If the purine photoproducts do form, they have negligible yields in comparison to pyrimidine photoproduct hence, the lack of need to study them.

Although it is highly unlikely that purine photoproducts will occur and cause damage to the DNA structure, it is possible to derive the structures that can form. This first type is a dimeric intermediate between two adjacent adenines that can form via a [2+2] Cycloaddition¹² which can undergo further arrangement reactions (Figure 11). The second is known as the ‘Porschke’ photoproduct and it can be formed from the ring opening of Adenine photoreaction intermediates (Figure 11)¹².The two products that can form after have been proposed after studies on adenine oligonucleotides⁵⁶ however there is little evidence to suggest a definite mechanism as well as the influence that causes them to form from the same intermediate.

2.4.           Oxidative Damage to DNA

It can be established that CPD’s are highly characteristic of UVB radiation. This leaves us to question, what role does UVA have in the prevalence of skin cancer? It can be concurred from several papers that occurrence of T<>T is decreased when DNA is exposed to UVA but as experiments indicate that it is more likely that UVA photons are absorbs by photosensitizers that in turn damage DNA via photo-oxidation reactions¹¹. UVA photons penetrate deeply into the dermis⁸², starting a chain of oxidative processes⁸³. This is a clear contrast to UVB, which has been shown only to be absorbed by components in the epidermal layer of the skin. Photodynamic effects can arise from endogenous species such as; cytochromes, Flavin and NAD(P)H; can react with DNA after UVA absorption via two types of mechanism, as discussed below.

In order to strike a balance, it is only fits that the gas that brings us life, can also bring out detrimental chaos. In its active forms, from by-products and intermediates of aerobic metabolic reactions, they can react with cellular components such as DNA, causing a cascade of reactions that can permanently damage the complex fabric of our genetic code.  As defined before, photosensitization mechanisms are described as the absorption of UV wavelengths by endogenous species. Upon absorption exogenous sensitisers are excited to their triplet excited state and can either return to the ground state, via decay mechanisms, or react with cellular material¹². These excited sensitisers can react via two main reaction pathways:

Type I – where an electron transfer and/or a Hydrogen abstraction occurs to yield free radicals.

Type II – where an energy transfer from a molecule that has absorbed UV light, to a molecular Oxygen. From this, the excited state Oxygen species, Singlet Oxygen, is generated of which is a very powerful oxidant.

Singlet Oxygen occupies the first excited singlet state (¹D_g) of molecular Oxygen¹². In this state, there are two electrons which occupy the same molecular orbital with paired spins. Its lifetime is in the microsecond (ms) scale which is relatively lengthy⁸⁴. There is the possibility of other excited states forming but, due to their higher energy and shorter lifetimes they often decay before a reaction can occur¹². It is for this reason that ¹D_g (which is referred to as Singlet Oxygen) takes the spotlight as the one of most relevance in biology. The energy transfer of singlet Oxygen (¹O₂ ) often not highly efficient and both Type I and Type II reactions occur simultaneously at a competitive rate to one another. It is for this reason when efforts to investigate the occurrence of  Reactive Oxygen Species (ROS) in DNA that both ¹O₂ and O₂^·– are found¹².

A common product of these reactions is 8-oxoGua lesions. While both UVB and UVA cannot provide enough energy to facilitate the one electron oxidation of the DNA bases but, due to the double stranded nature of DNA structure, the ionization potential is reduced, providing an alternative path for the reaction. UVA-mediated oxidation reactions to cellular DNA are started by the interaction between endogenous chromophores in cells and UV longwave radiation². This hypothesis is under constant reconstruction as the exact mechanisms and photosensitizers involved remain unclear⁸⁵.

2.4.1.      Reactive Oxygen Species

Extensive reports which detail investigations to determine the origins and mechanisms of ROS^65,86,87. Cadet et al provide one of the best hypothesis  as they outline several mechanisms by which ROS can form and give insight into some of the reactions that can occur between ROS and DNA. Both UVA and UVB can induce redox processes in cells as they both reach the basal cell layer within the epidermis of skin⁸⁸. UVA can also penetrate into deeper layers of the skin, and this could be relevant as a link to skin carcinogenesis. ROS that has been generated by UVB have been well documented, and a range of techniques have been used to detect the formation of Hydrogen Peroxide(H₂O₂), Superoxide anion (O₂^·–) and other species⁸⁶. When different UV wavelengths penetrate the skin, reactive species are not instantly generated as they are delayed from forming by the activation of enzymes which can later provide protection as antioxidants⁸⁹.

Eventually these wavelengths are absorbed by non-DNA chromophores (as well as DNA to obtain photoproducts detailed previously) which form photo excited states, like DNA, in which ROS can be generated as well as reactive nitrogen species (RNS),  organic free radicals and other toxic photoproducts. Human skin is abundant in non-DNA chromophores such as flavins, tryptophan, porphyrins and NADPH, which can absorb UVA and visible blue light⁸⁵.

Free radicals can be produced  by cells as a product of cellular metabolisms or exogenous species⁹⁰. This provides us with two types of species to consider when answering the question; “What can trigger a photo induced reaction on the nitrogenous bases of DNA?”. There are several species, one of them being the hydroxyl radical (^•OH), that can be used to characterise ‘oxidative damage to DNA’ and we can collectively call them Reactive Oxygen Species (ROS). ROS are notorious for their affiliation with mutagenesis, carcinogenesis and ageing⁹⁰, providing us with good footing into understanding UVA damage further. ROS can bring about all kinds of DNA damage; base modification, strand breakage, and DNA-Protein cross links to name a few with the products formed being dependant on the type of reaction, and the species involved.

2.4.2.      Type 1 Oxidation One Electron oxidation of nucleobases

This is the first possible pathway that can occur and it takes place via a Type I photosensitization mechanism. The efficiency of this mechanism of driven by the oxidation potential which acts favourably on Guanine. Many studies have been performed, on both isolated bases and nucleosides, where the target has been exposed to UVA radiation and/or visible light with specific photosensitizers present. From these reactions, products for all four of the DNA bases have been isolated and charcterized⁸². It has been made clear from such studies that Guanine is the main target as it has the lowest ionisation potential.

When Guanine loses an electron, the radical cation Gua^·+ is formed and it can undergo a variety of reactions. In the presence of reducing conditions, the radical can be converted into FapyGua (Figure 12), but when oxidized, its generates 8-oxo-Gua (Figure 12). Moreover, deprotonation reactions can give rise to Gua(-H)^·, which in turn can lead to formation of imidazole derivatives⁸². Further to this the radical cation can undergo a series of reactions, which leads to the generation of Oxazalone rearrangement products (Figure 12)¹¹. These products can be used to investigate other types of DNA lesions, such as DNA-protein crosslinks, but they will not be used in this project. It is also possible that these same reactions can occur on adenine which produces products such as FapyAde and 8-oxo-Ade occurring after the radical cation, Ade^·+, has been generated⁹¹. It is not much use investigating these, as this pathway is only minor.

Pyrimidine type I reactions are not often witnessed but there are radical cations created from their bases. The most occurring reaction these radicals undertake is when the C6 atom of a pyrimidine base is hydrated, which gives a carbon centred radial at the C5 atom¹¹. In Oxygen, this radical can generate hydroperoxyl radicals and hydroperoxides, which can go on to react further.

A reaction with molecular Oxygen yields the superoxide anion radical (O₂^•-)⁸²  and it can be highly oxidizing⁹². What results is a peroxyl radicals (HOO^·) or even hyper peroxides which have important intermediates to facilitate the formation of final oxidation products, which in the case of DNA, exhibit deleterious properties disrupting the production of vital amino acids and in turn, Proteins which ends the reaction chain with carcinogenic consequences. Normally, the radical ions formed by the photosensitizer are oxidized back to their starting molecules by O₂ that is present in the reaction environment but this in turn releases O₂^•-. Apart from Guanine, O₂^•- does not react with any of the other DNA bases but, it can spontaneously react to form peroxide (H₂O₂) which is also classified as a ROS. H₂O₂ can react in the Fenton reaction to generate ^•OH which is highly reactive and can effectively add to the double bonds of DNA bases or abstract hydrogen atoms⁹³.

2.4.3.      Hydroxyl radical-mediated DNA damage

When understanding UV genotoxicity the hydroxyl radical (^·OH) must be considered because although photochemical reactions do not directly initiate it, it provides a third oxidative pathway. ^·OH is a highly reactive species and plays an important role in many radiobiology studies⁶⁵.It reacts efficiently at the C5-C6 double bond, at diffusion controlled rates of 3-10 x10⁹  M^-1s^{-1 93}. This gives rise to ^·OH-mediated methyl oxidation products, In addition to this it can successfully react with moieties of 2-deoxyribose sugars which, in several proposed pathways, leads to the formation of DNA strand breaks⁹⁴.

 In UVA radiation, ^·OH it is thought to be generated from the fate of a far less reactive O₂^•-, which as discussed above, is produced in a type I photosensitization reaction when the photosensitizer radical anion reacts with molecular Oxygen. O₂^•- can also be released by mitochondria, as a response to UVA radiation⁹⁵. When O₂^•- interacts with H₂O₂, the Fenton reaction leads to the formation of ^·OH. ^·OH is not specific to a certain base like the other two reactants discussed and it reacts with all DNA components, including the sugar-phosphate backbone, at diffusion controlled rates⁹⁶. Reactions of ^·OH with the purine bases generates the transient formation of C8 hydroxylated radical which leads to products 8-oxoGua and FapyGua (Figure 12).

2.4.4.      Singled Oxygen Mediated Oxidation of Guanine

The  second type of photosensitization pathway occurs,  results from an energy transfer by the sensitiser in the triplet excited state to molecular Oxygen (O₂), which becomes a singlet species (¹D_g) that is reactive. This species tends to exhibit a powerful likeness for molecules that are rich in double bonds and the products that form are often dioxetanes, endoperoxides or end-oxidations. Singlet Oxygen (¹O₂) only reacts efficiently with Guanine, which is converted into a 4,8-endoperoxide when it undergoes a ¹O₂  – mediated Diels-Alder[4+2] photocycloaddition. In double stranded DNA, the intermediate of this reaction rearranges into 8-oxoGua (Figure 12) which is a marker for guanine oxidation and the major product^97,98. By treating a cell culture with endoperoxides, Ravenat et al⁹⁸ were able to evaluate this product by measuring the quantity of this product at different temperatures and detecting it using HPLC-MS/MS methods. This experiment only confirmed the direct oxidation of the guanine moiety in DNA and they suggested that the origin of singlet Oxygen is likely to be metabolic processes.

It is worth noting at this point that although it is a main product of UV induced Oxidation, 8-oxoGua is not a major mutation because it is thought to induce GCàTA transversions which are not considered to be a cause of major concern¹². This is concordant with the hypothesis that pyrimidine dimers are the main cause of mutations in DNA that lead to skin cancer.

3.     Conclusion and Outlook

UVB radiation is answerable for most of the photoproducts that occur from the direct absorption of UV photons by DNA and it is no surprise that previously most sun-creams provided protection only against UVB. The sharp increase in evidence that UVA is no innocent contributor to the formation of mutagenic lesions has shed light in the risk that UV exposure poses, increasing the demand for sun-cream that protects from both UV categories as well as a need to further study the methods of product formation. What can be drawn from previous experiments is that cellular DNA can undergo effective non-radiative decay mechanisms to the ground state. This agrees with the hypothesis that DNA is photostable along with distortion in the structure which allows dissipation of vibrational heat when the excimer returns to its ground state. The argument that DNA is photostable can also be supported by the fact that there are mechanisms in place for the most common photoproduct but this is only to as not all lesions are correctly identified, and mutations corrected. This can only conclude that to a certain extent DNA photostability allows for photoreactions to occur, and new techniques of measurement and analysis could provide vital insight in filling gaps in excited state mechanism theory  that would greatly contribute to the long term goal in understanding this subject.

Directing a sole focus in understanding excited states of DNA mechanisms and photoreaction pathways will not solve the imminent problem of skin cancer other harmful effects of UV it is clear that DNA is photostable. Although, It may help to inspire creative and innovative ways to fix lesions that do escape the notice of current defence mechanism, which is already proving their worth to protects us from damage. By looking at plant species and bacteria which have proven to have far more advanced photo-defence mechanisms then perhaps there is the possibility of providing a breakthrough in skin cancer prevention or even treatment.

References

1. P. G. UNNA, Br. J. Dermatol., 1895, 7, 83–85.
2. J. Cadet, E. Sage and T. Douki, Mutat. Res. Mol. Mech. Mutagen., 2005, 571, 3–17.
3. F. El Ghissassi, R. Baan, K. Straif, Y. Grosse, B. Secretan, V. Bouvard, L. Benbrahim-Tallaa, N. Guha, C. Freeman, L. Galichet and V. Cogliano, Lancet Oncol., 2009, 10, 751–752.
4. L. Marrot and J. R. Meunier, J. Am. Acad. Dermatol., , DOI:10.1016/j.jaad.2007.12.007.
5. R. K. Sivamani, L. A. Crane and R. P. Dellavalle, Dermatol. Clin., 2009, 27, 149–154.
6. R. P. Rastogi, Richa, A. Kumar, M. B. Tyagi and R. P. Sinha, J. Nucleic Acids, 2010, 2010, 1–32.
7. M. Blackburn and M. Gait, eds. G. M. Blackburn, M. J. Gait, D. Loakes and D. M. Williams, Royal Society of Chemistry, Cambridge, 2nd edn., 2007, pp. 316–322.
8. A. Stapleton, Plant Cell, 1992, 4, 1353–1358.
9. A. P. Schuch, N. C. Moreno, N. J. Schuch, C. F. M. Menck and C. C. M. Garcia, Free Radic. Biol. Med., 2017, 107, 110–124.
10. F. R. De Gruijl, in Skin Pharmacology and Applied Skin Physiology, 2002, vol. 15, pp. 316–320.
11. J.-L. Ravanat, T. Douki and J. Cadet, J. Photochem. Photobiol. B Biol., 2001, 63, 88–102.
12. D. I. Pattison and M. J. Davies, in Cancer: Cell Structures, Carcinogens and Genomic Instability, Birkhäuser-Verlag, Basel, 2006, pp. 131–157.
13. M. Barbatti, A. C. Borin and S. Ullrich, Top. Curr. Chem., 2015, 355, 1–32.
14. S. Marguet and D. Markovitsi, J. Am. Chem. Soc., 2005, 127, 5780–5781.
15. J. D. WATSON and F. H. C. CRICK, Nature, 1969, 224, 470–471.
16. C. Fonseca Guerra, F. M. Bickelhaupt, J. G. Snijders and E. J. Baerends, Chem. – A Eur. J., 1999, 5, 3581–3594.
17. W. J. Schreier, P. Gilch and W. Zinth, Annu. Rev. Phys. Chem., 2015, 66, 497–519.
18. P. Mignon, S. Loverix, J. Steyaert and P. Geerlings, Nucleic Acids Res., 2005, 33, 1779–1789.
19. C. E. Crespo-Hernández, B. Cohen and B. Kohler, Nature, 2005, 436, 1141–1144.
20. C. E. Crespo-Hernández, B. Cohen, P. M. Hare and B. Kohler, Chem. Rev., 2004, 104, 1977–2019.
21. K. Kleinermanns, D. Nachtigallová and M. S. de Vries, Int. Rev. Phys. Chem., 2013, 32, 308–342.
22. C. T. Middleton, K. de La Harpe, C. Su, Y. K. Law, C. E. Crespo-Hernández and B. Kohler, Annu. Rev. Phys. Chem., 2009, 60, 217–239.
23. J. Peon and A. H. Zewail, Chem. Phys. Lett., 2001, 348, 255–262.
24. J.-M. L. Pecourt, J. Peon and B. Kohler, J. Am. Chem. Soc., 2001, 123, 10370–10378.
25. in IUPAC Compendium of Chemical Terminology, IUPAC, Research Triagle Park, NC.
26. D. Voet, W. B. Gratzer, R. A. Cox and P. Doty, Biopolymers, 1963, 1, 193–208.
27. L. Serrano-Andrés and M. Merchán, J. Photochem. Photobiol. C Photochem. Rev., 2009, 10, 21–32.
28. W. J. Schreier, P. Gilch and W. Zinth, Annu. Rev. Phys. Chem., 2015, 66, 497–519.
29. J.-M. L. Pecourt, J. Peon and B. Kohler, J. Am. Chem. Soc., 2000, 122, 9348–9349.
30. S. Ullrich, T. Schultz, M. Z. Zgierski and A. Stolow, Phys. Chem. Chem. Phys., 2004, 6, 2796.
31. C. Canuel, M. Mons, F. Piuzzi, B. Tardivel, I. Dimicoli, C. Canuel, M. Mons, F. Piuzzi and B. Tardivel, J. Chem. Phys. J. Chem. Phys. J. Chem. Phys. J. Chem. Phys. J. Chem. Phys. J. Chem. Phys. J. Chem. Phys., , DOI:10.1063/1.1850469.
32. N. Ismail, L. Blancafort, M. Olivucci, B. Kohler and M. A. Robb, J. Am. Chem. Soc., 2002, 124, 6818–6819.
33. C. M. Marian, J. Chem. Phys., 2005, 122, 104314.
34. L. Serrano-Andrés, M. Merchán and A. C. Borin, J. Am. Chem. Soc., 2008, 130, 2473–2484.
35. H. R. Hudock, B. G. Levine, A. L. Thompson, H. Satzger, D. Townsend, N. Gador, S. Ullrich, A. Stolow and T. J. Martínez, J. Phys. Chem. A, 2007, 111, 8500–8508.
36. H. R. Hudock and T. J. Martínez, ChemPhysChem, 2008, 9, 2486–2490.
37. J. J. Szymczak, M. Barbatti, J. T. Soo Hoo, J. A. Adkins, T. L. Windus, D. Nachtigallová and H. Lischka, J. Phys. Chem. A, 2009, 113, 12686–12693.
38. M. Barbatti and H. Lischka, J. Am. Chem. Soc., 2008, 130, 6831–6839.
39. M. Barbatti, A. J. A. Aquino, J. J. Szymczak, D. Nachtigallova, P. Hobza and H. Lischka, Proc. Natl. Acad. Sci., 2010, 107, 21453–21458.
40. V. Bonačić-Koutecký, J. Koutecký and J. Michl, Angew. Chemie Int. Ed. English, 1987, 26, 170–189.
41. D. B. Bucher, A. Schlueter, T. Carell and W. Zinth, Angew. Chemie Int. Ed., 2014, 53, 11366–11369.
42. A. L. Sobolewski and W. Domcke, Chem. Phys., 2003, 294, 73–83.
43. T. Schultz, Science (80-. )., 2004, 306, 1765–1768.
44. S. Perun, A. L. Sobolewski and W. Domcke, J. Phys. Chem. A, 2006, 110, 13238–13244.
45. C. E. Crespo-Hernández, B. Cohen and B. Kohler, Nature, 2005, 436, 1141–1144.
46. T. Takaya, C. Su, K. de La Harpe, C. E. Crespo-Hernandez and B. Kohler, Proc. Natl. Acad. Sci., 2008, 105, 10285–10290.
47. S. Takeuchi and T. Tahara, Proc. Natl. Acad. Sci., 2007, 104, 5285–5290.
48. H. Kang, K. T. Lee, B. Jung, Y. J. Ko and S. K. Kim, J. Am. Chem. Soc., 2002, 124, 12958–9.
49. J. D. Spikes, in The Science of Photobiology, Springer US, Boston, MA, 1989, pp. 79–110.
50. T. Matsunaga, K. Hieda and O. Nikaido, Photochem. Photobiol., 1991, 54, 403–410.
51. G. B. Bauer and L. F. Povirk, Nucleic Acids Res., 1997, 25, 1211–1218.
52. R. González-Luque, T. Climent, I. González-Ramírez, M. Merchán and L. Serrano-Andrés, J. Chem. Theory Comput., 2010, 6, 2103–2114.
53. T. Douki, M. Court, S. Sauvaigo, F. Odin and J. Cadet, .
54. R. Beukers, A. P. M. Eker and P. H. M. Lohman, DNA Repair (Amst)., 2008, 7, 530–543.
55. C. E. Crespo-Hernández, B. Cohen and B. Kohler, Nature, 2005, 436, 1141–1144.
56. R. O. Rahn, Nucleic Acids Res., 1976, 3, 879–890.
57. V. I. Danilov, O. N. Slyusarchuk, J. L. Alderfer, J. J. P. Stewart and P. R. Callis, Photochem. Photobiol., 1994, 59, 125–129.
58. T. Douki and J. Cadet, Biochemistry, 2001, 40, 2495–2501.
59. T. Douki, Photochem. Photobiol. Sci., 2013, 12, 1286.
60. W. A. Korfmacher, Drug Discov. Today, 2005, 10, 1357–1367.
61. W. M. Kwok, C. Ma and D. L. Phillips, J. Am. Chem. Soc., 2008, 130, 5131–5139.
62. E. Ben-Hur and R. Ben-Ishai, Biochim. Biophys. Acta – Nucleic Acids Protein Synth., 1968, 166, 9–15.
63. R. B. Setlow, W. L. Carrier and F. J. Bollum, Proc. Natl. Acad. Sci. U. S. A., 1965, 53, 1111–8.
64. W. Peng and B. R. Shaw, Biochemistry, 1996, 35, 10172–10181.
65. J. Cadet, S. Mouret, J.-L. Ravanat and T. Douki, Photochem. Photobiol., 2012, 88, 1048–1065.
66. J. S. Taylor and M. P. Cohrs, J. Am. Chem. Soc., 1987, 109, 2834–2835.
67. B. M. Pilles, D. B. Bucher, L. Liu, P. Clivio, P. Gilch, W. Zinth and W. J. Schreier, J. Phys. Chem. Lett., 2014, 5, 1616–1622.
68. M. D’Auria and R. Racioppi, Molecules, 2013, 18, 11384–11428.
69. A. Banyasz, T. Douki, R. Improta, T. Gustavsson, D. Onidas, I. Vayá, M. Perron and D. Markovitsi, J. Am. Chem. Soc., 2012, 134, 14834–14845.
70. K. Haiser, B. P. Fingerhut, K. Heil, A. Glas, T. T. Herzog, B. M. Pilles, W. J. Schreier, W. Zinth, R. de Vivie-Riedle and T. Carell, Angew. Chemie Int. Ed., 2012, 51, 408–411.
71. Y.-H. You, D.-H. Lee, J.-H. Yoon, S. Nakajima, A. Yasui and G. P. Pfeifer, J. Biol. Chem., 2001, 276, 44688–44694.
72. J. Jans, W. Schul, Y.-G. Sert, Y. Rijksen, H. Rebel, A. P. M. Eker, S. Nakajima, H. van Steeg, F. R. de Gruijl, A. Yasui, J. H. J. Hoeijmakers and G. T. J. van der Horst, Curr. Biol., 2005, 15, 105–115.
73. M. M. Valejo Coelho, T. R. Matos and M. Apetato, Clin. Dermatol., 2016, 34, 563–570.
74. Z. Liu, C. Tan, X. Guo, Y.-T. Kao, J. Li, L. Wang, A. Sancar and D. Zhong, Proc. Natl. Acad. Sci., 2011, 108, 14831–14836.
75. H. W. Park, S. T. Kim, A. Sancar and J. Deisenhofer, Science, 1995, 268, 1866–72.
76. A. Sancar, Chem. Rev., 2003, 103, 2203–2238.
77. Y.-T. Kao, C. Saxena, L. Wang, A. Sancar and D. Zhong, Proc. Natl. Acad. Sci., 2005, 102, 16128–16132.
78. N. M. Wikonkal and D. E. Brash, J. Investig. dermatology. Symp. Proc., 1999, 4, 6–10.
79. T. M. Rünger and U. P. Kappes, Photodermatol. Photoimmunol. Photomed., 2008, 24, 2–10.
80. N. Dumaz, C. Drougard,  a Sarasin and L. Daya-Grosjean, Proc. Natl. Acad. Sci. U. S. A., 1993, 90, 10529–10533.
81. H. Nakazawa, D. English, P. L. Randell, K. Nakazawa, N. Martel, B. K. Armstrong and H. Yamasaki, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 360–364.
82. J. Cadet, T. Douki, J.-L. Ravanat and P. Di Mascio, Photochem. Photobiol. Sci., 2009, 8, 903.
83. R. . Tyrrell, in Oxidative Stress: Oxidants and Antioxidants, ed. H. Sies, Academic Press, London, 1990, pp. 57–83.
84. M. J. Davies, Biochem. Biophys. Res. Commun., 2003, 305, 761–770.
85. G. T. Wondrak, M. K. Jacobson and E. L. Jacobson, Photochem. Photobiol. Sci., 2006, 5, 215–237.
86. D. E. Heck, A. M. Vetrano, T. M. Mariano and J. D. Laskin, J. Biol. Chem., 2003, 278, 22432–22436.
87. J. Cadet, M. Berger, T. Douki and J.-L. Ravanat, Rev. Physiol. Biochem. Pharmacol. Vol. 131, 1997, 1–87.
88. E. Sage, P.-M. Girard and S. Francesconi, Photochem. Photobiol. Sci., 2012, 11, 74–80.
89. G.-H. Jin, Y. Liu, S.-Z. Jin, X.-D. Liu and S.-Z. Liu, Radiat. Environ. Biophys., 2007, 46, 61–68.
90. B. Halliwell and J. M. . Gutteridge, Free radicals in Biology and Medicine, OUP Oxford, Oxford, 5th edn., 2015.
91. S. Steenken, Chem. Rev., 1989, 89, 503–520.
92. J. Cadet, T. Douki and J.-L. Ravanat, Nat. Chem. Biol., 2006, 2, 348–349.
93. M. Dizdaroglu, P. Jaruga, M. Birincioglu and H. Rodriguez, Free Radic. Biol. Med., 2002, 32, 1102–1115.
94. W. K. Pogozelski and T. D. Tullius, Chem. Rev., 1998, 98, 1089–1108.
95. R. Gniadecki, T. Thorn, J. Vicanova, A. Petersen and H. C. Wulf, J. Cell. Biochem., 2001, 80, 216–222.
96. J. Cadet, T. Delatour, T. Douki, D. Gasparutto, J.-P. Pouget, J.-L. Ravanat and S. Sauvaigo, Mutat. Res. Mol. Mech. Mutagen., 1999, 424, 9–21.
97. J. Cadet, J.-L. Ravanat, G. R. Martinez, M. H. G. Medeiros and P. Di Mascio, Photochem. Photobiol., 2006, 82, 1219.
98. J.-L. Ravanat, P. Di Mascio, G. R. Martinez, M. H. G. Medeiros and J. Cadet, J. Biol. Chem., 2000, 275, 40601–40604.