Why does sun damage to DNA continue hours after exposure?

Acta Eruditorum

Abby Van Voorhees

Dr. Van Voorhees is the physician editor of Dermatology World. She interviews the author of a recent study each month.

Bookmark and Share

In this month’s Acta Eruditorum column, Physician Editor Abby S. Van Voorhees, MD, talks with Douglas Brash, PhD, about his recent Science article, "Chemiexcitation of melanin derivatives induces DNA photoproducts long after UV exposure."

Dr. Van Voorhees: What do we know about the mutations that are seen in sun-induced melanomas? What is the UV signature?

Dr. Brash: Melanomas on sun-exposed skin have acquired mutations in many genes. Most of these mutations are “UV signature” mutations — cytosine mutating to a thymine at a location where cytosine’s neighbor was a thymine or another cytosine. The only carcinogen that makes this kind of mutation frequently is ultraviolet light, implicating sunlight as the mutagen.

Ultraviolet light also makes less specific mutations, with about a quarter of its mutations being indistinguishable from those made by other carcinogens. So if we see UV signature mutations in a tumor, then many of the other mutations must have been caused by UV too. For example, the famous BRAF mutation in melanoma is not a UV signature mutation but it might yet have been caused by UV; we just can’t tell.

Dr. Van Voorhees: Tell us about CPDs. What cell are they derived from? What about the timing of when they are formed? The cell of origin? Are they created by exposure to both UVB and UVA? Is there a genetic disorder that closely correlates?

Dr. Brash: When two neighboring pyrimidine bases (that is, cytosine or thymine) absorb a UV photon, their double bonds rearrange within a picosecond and join the two bases together. The resulting cyclobutane pyrimidine dimer (CPD) puts a bend in the DNA that makes it difficult for the cell to copy its DNA correctly. When a person goes to the beach, many CPDs are created in every cell of the skin. Fortunately, a DNA repair system called excision repair removes most of the CPDs and replaces them with normal DNA. Xeroderma pigmentosum patients lack one or another of the genes required for excision repair, so children with XP have a 10,000-fold greater incidence of skin cancer. UVB is much better than UVA at making CPDs in purified DNA, but UVA penetrates deeper into the skin and there is 20 times more UVA in sunlight.

Dr. Van Voorhees: Are some CPDs more important in causing this genetically altered signature?

Dr. Brash: The most frequent CPD is the one that joins two thymines, a “TT CPD.” It causes cell death. But the UV signature mutations are caused by the less frequent TC and CT CPDs.[pagebreak]

Dr. Van Voorhees: Explain to us what you learned about how this occurs scientifically in melanin-containing cells.

Dr. Brash: In cells containing melanin, the events described above still happen, although about half as often because the melanin acts as a shield by absorbing much of the UV. As we expected, CPDs could be measured as soon as the UVA or UVB lamp was turned off. The surprise was that, when melanin was present, CPDs continued to be made for another three to four hours — ultimately as many or more than were created initially by the standard route.

What happens turned out to be a three-part story. First, UV activates two enzymes; one normally makes an oxygen free radical, superoxide; the other makes a nitrogen free radical, nitric oxide, that is widely used in the cell as a signaling molecule. This is the slow step that can continue for several hours. These reactive molecules combine to make peroxynitrite, a very strong oxidizing agent that initiates a chemical reaction that ultimately excites an electron in a fragment of melanin. Most chemical reactions in the cell instead increase a molecule’s vibrational energy so that it can react with other molecules. The electron reaches a high energy usually only possible after ultraviolet radiation exposure. If DNA happens to be nearby, this energy is transferred to the DNA in the dark, creating the same CPD that sunlight causes in daylight. Chemically induced electron excitation is called “chemiexcitation” and was previously encountered only in bioluminescent lower organisms such as jellyfish and fireflies.

Why would nature do this? There seems to be a balance between “melanin is good” and “melanin is bad,” with dark eumelanin having a better ratio than yellow pheomelanin. There might be a perfect sunlight absorber out there, but melanin is the best that evolution came up with. It is better than no melanin at all, particularly at young ages because evolution does not care about cancers that arise after the child-bearing years. The imperfect melanin strategy does serve to spread the CPDs out over time, which may be better than occurring all at once during the sun exposure and possibly overwhelming the DNA repair system.

Dr. Van Voorhees: What about in albino skin — how does this mechanism apply in this situation?

Dr. Brash: Albino individuals have a higher frequency of actinic keratoses and non-melanoma skin cancer, so melanin has a net benefit. It is less clear to me what the melanoma incidence is in albinos. Published studies may not have examined enough albino and matched pigmented individuals to detect a difference unless it were an increase and were greater than seen for non-melanoma skin cancer.

Dr. Van Voorhees: What is the implication of this work to the practicing dermatologist? Is melanin both protective and potentially carcinogenetic? Does this work suggest possible future treatments to prevent melanomas?

Dr. Brash: While this may have been evolution’s best solution, it doesn’t have to be ours. The delayed pathway should be interceptable at several points: preventing enzyme activation, scavenging free radicals and peroxynitrite, and diverting the energy from the excited electron into heat before it can transfer to DNA.

Dr. Brash is professor of therapeutic radiology and dermatology at Yale School of Medicine. His article appeared in the February 20, 2015 issue of Science. Vol. 347 no. 6224 pp. 842-847 DOI: 10.1126/science.1256022.